Peer-Review Draft: Report on Carcinogens Monograph on Cobalt and Certain Cobalt Compounds

Peer-Review Draft:
Report on Carcinogens Monograph on
Cobalt and Certain Cobalt Compounds
June 5, 2015
Office of the Report on Carcinogens
Division of the National Toxicology Program
National Institute of Environmental Health Sciences
U.S. Department of Health and Human Services
This information is distributed solely for the purpose of pre-dissemination peer review under applicable
information quality guidelines. It has not been formally distributed by the National Toxicology Program.
It does not represent and should not be construed to represent any NTP determination or policy.
This Page Intentionally Left Blank
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Foreword
The National Toxicology Program (NTP) is an interagency program within the Public Health
Service (PHS) of the Department of Health and Human Services (HHS) and is headquartered at
the National Institute of Environmental Health Sciences of the National Institutes of Health
(NIEHS/NIH). Three agencies contribute resources to the program: NIEHS/NIH, the National
Institute for Occupational Safety and Health of the Centers for Disease Control and Prevention
(NIOSH/CDC), and the National Center for Toxicological Research of the Food and Drug
Administration (NCTR/FDA). Established in 1978, the NTP is charged with coordinating
toxicological testing activities, strengthening the science base in toxicology, developing and
validating improved testing methods, and providing information about potentially toxic
substances to health regulatory and research agencies, scientific and medical communities, and
the public.
The Report on Carcinogens (RoC) is prepared in response to Section 301 of the Public Health
Service Act as amended. The RoC contains a list of identified substances (i) that either are
known to be human carcinogens or are reasonably anticipated to be human carcinogens and (ii)
to which a significant number of persons residing in the United States are exposed. The
Secretary, Department of HHS, has delegated responsibility for preparation of the RoC to the
NTP, which prepares the report with assistance from other Federal health and regulatory
agencies and nongovernmental institutions. The most recent RoC, the 13th Edition (2014), is
available at http://ntp.niehs.nih.gov/go/roc13.
Nominations for (1) listing a new substance, (2) reclassifying the listing status for a substance
already listed, or (3) removing a substance already listed in the RoC are evaluated in a scientific
review process (http://ntp.niehs.nih.gov/go/rocprocess) with multiple opportunities for scientific
and public input and using established listing criteria (http://ntp.niehs.nih.gov/go/15209). A list
of candidate substances under consideration for listing in (or delisting from) the RoC can be
obtained by accessing http://ntp.niehs.nih.gov/go/37893.
This draft document should not be construed to represent final NTP determination or policy
i
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Background and Methods
Cobalt is a naturally occurring element that is present in several different forms. Elemental
cobalt is a hard, silvery grey metal that can combine with other elements, e.g., with oxygen
(cobalt oxide), sulfur (cobalt sulfate) or arsenic (cobalt arsenide). The most common oxidation
states of cobalt are +2 and +3; for most simple cobalt compounds, the valence is +2, designated
as cobalt(II). Cobalt compounds can be organic or inorganic as well as water-soluble or insoluble. Cobalt compounds are used in a variety of industrial applications and as a colorant for
glass, ceramics, and paint, and as catalysts, as driers for inks and paints, and in feed supplements
and batteries. Cobalt is used in alloys or composites, such as cobalt-tungsten carbide, and in
cobalt-containing prosthetics. Cobalt nanoparticles are used in medical tests and treatments as
well as in the textile and electronics industries.
‘Cobalt and certain cobalt compounds’ was selected for review for possible listing in the Report
on Carcinogens (RoC) based on evidence of widespread exposure and an adequate database of
cancer studies to evaluate the potential carcinogenicity of cobalt. “Certain” refers to those cobalt
compounds that release cobalt ions in vivo, which does not include cobalt as part of the vitamin
B12 molecule because of the stability of that molecule in biological fluids. Cancer and toxicology
studies of forms of cobalt that have confounding exposures, such as cobalt alloys and radioactive
forms of cobalt, were not included in the review of cobalt and certain cobalt compounds. Two
cobalt-containing substances, ‘cobalt sulfate’ and ‘cobalt-tungsten carbide: powders and hard
metals,’ are currently listed in the Report on Carcinogens (RoC) as reasonably anticipated to be
human carcinogens (NTP 2014d, 2014a). Cobalt sulfate, which has been listed since 2004 based
on sufficient evidence of carcinogenicity from studies in experimental animals (NTP 2002b), is
included in the current review of cobalt as a class. Cobalt-tungsten carbide: powders and hard
metals, which was first listed in 2011 based on limited evidence of carcinogenicity from studies
in humans and supporting evidence from studies on mechanisms of carcinogenesis (NTP 2009)
falls outside the review.
Monograph contents
This RoC draft monograph on cobalt consists of the following components: (Part 1) the cancer
evaluation component that reviews the relevant scientific information and assesses its quality,
applies the RoC listing criteria to the scientific information, and recommends an RoC listing
status for cobalt and certain cobalt compounds, and (Part 2, which will be drafted with input
from the internal review group) the draft substance profile containing the NTP’s preliminary
listing recommendation, a summary of the scientific evidence considered key to reaching that
recommendation, and data on properties, use, production, exposure, and Federal regulations and
guidelines to reduce exposure to cobalt and certain cobalt compounds.
The methods for preparing the draft RoC monograph on cobalt and certain cobalt compounds are
described in the “Cobalt and Certain Cobalt Compounds Protocol”(NTP 2014c). The cancer
evaluation component for cobalt and certain cobalt compounds provides information on the
following topics that are relevant to understanding the relationship between exposure to cobalt
compounds and cancer: Introduction and properties (Section 1), human exposure (Section 2),
disposition and toxicokinetics (Section 3), human cancer studies (Section 4), studies in
experimental animals (Section 5), mechanisms and other relevant effects (Section 6), and an
ii
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
overall cancer evaluation that provides a synthesis of Sections 1 through 6 and rationale for
listing cobalt and certain cobalt compounds as a class (Section 7). The information reviewed in
Sections 3 through 7 (except for information on exposure and properties) must come from
publicly available, peer-reviewed sources. The appendices in the RoC Monograph contain
important supplementary information, such as the literature search strategy, study quality tables,
and study descriptions and results for some sections.
Process for preparation of the cancer hazard evaluation component
The process for preparing the cancer evaluation component of the monograph included
approaches for obtaining public and scientific input and using systematic methods (e.g.,
standardized methods for identifying the literature [see Appendix A], inclusion/exclusion
criteria, extraction of data and evaluation of study quality using specific guidelines, and
assessment of the level of evidence for carcinogenicity using established criteria). [Links are
provided within the document to the appendices, and specific tables or sections can be selected
from the table of contents.]
The Office of the Report on Carcinogens (ORoC) followed the approaches outlined in the
concept document, which discusses the scientific issues and questions relevant to the evaluation
of the carcinogenicity of cobalt and certain cobalt compounds, the scope and focus of the
monograph, and the approaches to obtain scientific and public input to address the key scientific
questions and issues for preparing the cancer evaluation component of the draft monograph. The
ORoC presented the concept document for cobalt to the NTP Board of Scientific Counselors
(BSC) at the April 17, 2014 meeting, which provided opportunity for written and oral public
comments, after which the concept was finalized and cobalt was approved by the NTP Director
as a candidate substance for review. The concept document is available on the RoC website
(http://ntp.niehs.nih.gov/go/730697).
Key scientific questions and issues relevant for the cancer evaluation
The scientific issues in this review concern the evaluation of the topics mentioned earlier,
including human exposure, disposition and toxicokinetics, cancer studies in humans and
experimental animals, and mechanistic data. The key questions for each topic are as follows:
Questions related to the evaluation of human exposure information
•
•
•
•
•
•
•
How are people in the United States exposed to cobalt and cobalt compounds?
How do we measure exposure?
What are the non-occupational sources and levels of exposure?
What are the occupational settings and levels of exposure?
Has exposure changed over time?
What federal regulations and guidelines limit exposure to cobalt?
Are a significant number of people residing in the United States exposed to cobalt and
cobalt compounds?
This draft document should not be construed to represent final NTP determination or policy
iii
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Questions related to the evaluation of disposition and toxicokinetics
•
•
•
•
How is cobalt absorbed, distributed, metabolized, and excreted (ADME)?
What, if any, are the qualitative and/or quantitative species or sex differences for ADME?
What is known about the form of cobalt (particulate, ion) from ADME studies in exposed
tissue, particularly in the lung?
How can toxicokinetic models (if any) inform biological plausibility, interspecies
extrapolation, or other mechanistic questions for cobalt?
Questions related to the evaluation of human cancer studies
•
•
•
•
•
Which epidemiologic studies should be included in the review?
What are the methodological strengths and limitations of these studies?
What are the potential confounders for cancer risk for the tumor sites of interest in these
studies?
Is there a credible association between exposure to cobalt and cancer?
If so, can the relationship between cancer endpoints and exposure to cobalt be explained
by chance, bias, or confounding?
Questions related to the evaluation of cancer studies in experimental animals
•
•
•
What is the level of evidence (sufficient or not sufficient) of carcinogenicity of cobalt
from animal studies?
What are the methodological strengths and limitations of the studies?
What are the tissue sites?
Questions related to the evaluation of mechanistic data and other relevant data
•
•
•
iv
What are the genotoxic effects due to cobalt exposure? Does genotoxicity vary by cobalt
compound?
What are the cytotoxic or toxic effects of cobalt exposure? Does cytotoxicity or toxicity
vary by cobalt compound?
What are the major mechanistic modes of action for the carcinogenicity of cobalt?
o What are the common key steps or mode(s) of action of toxicity or
carcinogenicity across different cobalt compounds? What role and contribution
does cobalt ion play in the proposed mechanism? What are the effects from
exposure to particulate cobalt?
o What factors influence biological or carcinogenic effects? How do particle size,
solubility, and cellular uptake of a cobalt compound affect biological or
carcinogenic effects?
o Is there evidence that supports grouping cobalt and certain cobalt compounds
together in the assessment?
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Approach for obtaining scientific and public input
To help address the approach to identify a common mode of action involving the cobalt ion for
certain cobalt compounds, additional scientific input was requested early in the review process to
define the scope of the review, i.e., what cobalt compound(s) could reasonably be included in
this evaluation? Based on input from several scientific experts at a Cobalt Information Group
Meeting convened at NIEHS on October 7, 2104, the scope of the evaluation was defined as
“cobalt and certain cobalt compounds,” where “certain” refers to those cobalt compounds that
release cobalt ions in biological fluids. Technical advisors for the review of cobalt and certain
cobalt compounds are identified on the “CONTRIBUTORS” page.
Public comments on scientific issues were requested at several times prior to the development of
the draft RoC monograph, including the request for information on the nomination, and the
request for comment on the draft concept document, which outlined the rationale and approach
for conducting the scientific review. In addition, the NTP posted its protocol for preparing the
draft RoC monograph on cobalt and certain cobalt compounds for public input on the ORoC
webpage for cobalt and certain cobalt compounds (http://ntp.niehs.nih.gov/go/730697) prior to
the release of the draft monograph. One written public comment on cobalt (in response to the
request for information on the nomination) has been received from the public as of the date on
this document.
Methods for writing the cancer evaluation component of the monograph
The procedures by which relevant literature was identified, data were systematically extracted
and summarized, and the draft monograph was written, together with the processes for scientific
review, quality assurance, and assessment and synthesis of data, are described below.
This draft document should not be construed to represent final NTP determination or policy
v
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
The preparation of the RoC monograph for cobalt and certain cobalt compounds began with
development of a literature search strategy to obtain information relevant to the topics listed
above for Sections 1 through 6 using search terms developed in collaboration with a reference
librarian (see Protocol). The 7,150 citations identified from these searches were uploaded to
web-based systematic review software for evaluation by two separate reviewers using
inclusion/exclusion criteria, and 465 references were selected for final inclusion in the draft
monograph using these criteria.
RoC Listing Criteria
Information for the relevant cancer and
mechanistic sections was systematically
extracted in tabular format and/or
summarized in the text, following
specific procedures developed by
ORoC, from studies selected for
inclusion in the monograph. All
sections of the monograph underwent
scientific review and quality assurance
(QA, i.e., assuring that all the relevant
data and factual information extracted
from the publications have been
reported accurately) by a separate
reviewer. Any discrepancies between
the writer and the reviewer were
resolved by mutual discussion in
reference to the original data source.
Strengths, weaknesses, and study
quality of the cancer studies for cobalt
compounds in humans (see Appendix
C) and experimental animals (see
Appendix D) were assessed based on a
series of a priori considerations
(questions and guidelines for answering
the questions), which are available in
the protocol (available at
http://ntp.niehs.nih.gov/go/730697).
Two reviewers evaluated the quality of
each study. Any disagreements between
the two reviewers were resolved by
mutual discussion or consultation with a
third reviewer in reference to the
original data source. Relevant
genotoxicity and mechanistic studies
were also assessed for their strengths
and weaknesses.
RoC listing criteria (see text box) were
vi
Known To Be Human Carcinogen:
There is sufficient evidence of carcinogenicity from studies
in humans*, which indicates a causal relationship between
exposure to the agent, substance, or mixture, and human
cancer.
Reasonably Anticipated To Be Human
Carcinogen:
There is limited evidence of carcinogenicity from studies in
humans*, which indicates that causal interpretation is
credible, but that alternative explanations, such as chance,
bias, or confounding factors, could not adequately be
excluded, OR
there is sufficient evidence of carcinogenicity from studies
in experimental animals, which indicates there is an
increased incidence of malignant and/or a combination of
malignant and benign tumors (1) in multiple species or at
multiple tissue sites, or (2) by multiple routes of exposure,
or (3) to an unusual degree with regard to incidence, site,
or type of tumor, or age at onset, OR
there is less than sufficient evidence of carcinogenicity in
humans or laboratory animals; however, the agent,
substance, or mixture belongs to a well-defined, structurally
related class of substances whose members are listed in a
previous Report on Carcinogens as either known to be a
human carcinogen or reasonably anticipated to be a human
carcinogen, or there is convincing relevant information that
the agent acts through mechanisms indicating it would
likely cause cancer in humans.
Conclusions regarding carcinogenicity in humans or
experimental animals are based on scientific judgment,
with consideration given to all relevant information.
Relevant information includes, but is not limited to, dose
response, route of exposure, chemical structure,
metabolism, pharmacokinetics, sensitive sub-populations,
genetic effects, or other data relating to mechanism of
action or factors that may be unique to a given substance.
For example, there may be substances for which there is
evidence of carcinogenicity in laboratory animals, but there
are compelling data indicating that the agent acts through
mechanisms which do not operate in humans and would
therefore not reasonably be anticipated to cause cancer in
humans.
*This evidence can include traditional cancer epidemiology studies,
data from clinical studies, and/or data derived from the study of
tissues or cells from humans exposed to the substance in question
that can be useful for evaluating whether a relevant cancer
mechanism is operating in people.
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
applied to the available database of carcinogenicity data to assess the level of evidence
(sufficient, limited, or inadequate) for the carcinogenicity of cobalt and certain cobalt compounds
from studies in humans and the level of evidence (sufficient, not sufficient) from studies in
experimental animals. The approach for synthesizing the evidence across studies and reaching a
level of evidence conclusion was outlined in the protocol. The evaluation of the mechanistic data
included a complete discussion and assessment of the strength of evidence for potential modes of
action for cobalt-induced neoplasia, including those involving, e.g., cytotoxicity, genotoxicity,
and oxidative stress. Mechanistic data are discussed across cobalt compounds. The RoC listing
criteria were then applied to the body of knowledge (cancer studies in humans and experimental
animals and mechanistic data) for cobalt and certain cobalt compounds to reach a listing
recommendation.
This draft document should not be construed to represent final NTP determination or policy
vii
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Contributors
Office of the Report on Carcinogens (ORoC), Division of the National Toxicology Program
(NTP)
Conducted technical review and evaluation and proposed the preliminary listing
recommendation
Ruth Lunn, DrPH
Director, ORoC
Gloria D. Jahnke, DVM, DABT
Diane L. Spencer, MS;
Co-Project Lead
Integrated Laboratory Systems, Inc. (Support provided through NIEHS Contract Number
HHSN273201100004C)
Conducted technical review and evaluation
Sanford Garner, PhD
Principal Investigator
Stanley Atwood, MS, DABT;
Co-Project Lead
Andrew Ewens, PhD, DABT
Pamela Schwingl, PhD
Alton Peters, MS
Provided administrative support
Ella Darden, BS
Tracy Saunders, BS
Other Members of the RoC Monograph Team (NIEHS/NTP except as noted)
Provided advice on defining the scope of review (i.e., served on the Information Group) or on
issues related to the cancer hazard evaluation
Mamta Behl, PhD, DABT
(Information Group Moderator)
Arun Pandiri, PhD, DACVP
Chad Blystone, PhD, DABT
Erik Tokar, PhD
Janet Carter, MPH (OSHA/DOL)
Michelle Hooth, PhD, DABT
viii
Matthew Stout, PhD, DABT
John Wheeler, PhD, DABT
(ATSDR/CDC)
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Part 1
Draft Cancer Hazard Evaluation
Properties and Chemical Identification
Human Exposure
Disposition (ADME) and Toxicokinetics
Human Cancer Studies
Studies of Cancer in Experimental Animals
Mechanistic Data and Other Relevant Effects
Overall Cancer Evaluation
This draft document should not be construed to represent final NTP determination or policy
ix
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
This Page Intentionally Left Blank
x
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Table of Contents
Part 1
1
Chemical identification and properties ..........................................................................1
1.1 Properties of cobalt metal and cobalt compounds, both soluble and poorly soluble
...............................................................................................................................1
1.2 Water solubility and bioaccessibility ....................................................................1
1.3 Variability of valence ............................................................................................4
1.4 Summary ...............................................................................................................4
2
Human Exposure............................................................................................................5
2.1 Mining and production ..........................................................................................5
2.2 Use .........................................................................................................................6
2.2.1 Metallurgical uses .....................................................................................7
2.2.2 Cemented carbides and bonded diamonds ................................................7
2.2.3 Chemical uses ...........................................................................................7
2.2.4 Electronics and “green” energy ................................................................8
2.3 Biomonitoring for cobalt .......................................................................................8
2.3.1 Evidence of exposure ................................................................................8
2.3.2 Methods...................................................................................................10
2.4 Characterization of exposure in the workplace ...................................................11
2.4.1 Hard-metal production, processing, and use ...........................................13
2.4.2 Cobalt-containing diamond tools and cobalt alloys ................................13
2.4.3 Pigments ..................................................................................................13
2.4.4 Cobalt production (metals and salts).......................................................13
2.5 Surgical implants .................................................................................................13
2.6 Other sources of exposure: Food, consumer and other medical products and tobacco
.............................................................................................................................14
2.7 Potential for environmental exposure..................................................................15
2.7.1 Releases...................................................................................................15
2.7.2 Occurrence ..............................................................................................16
2.7.3 Exposure .................................................................................................17
2.8 Summary and synthesis .......................................................................................17
3
Disposition and Toxicokinetics....................................................................................19
3.1 Disposition...........................................................................................................19
3.1.1 Humans ...................................................................................................20
3.1.2 Experimental animals..............................................................................22
3.2 Toxicokinetics .....................................................................................................28
3.2.1 Humans ...................................................................................................29
3.2.2 Experimental animals..............................................................................29
3.3 Synthesis ..............................................................................................................30
4
Human Cancer Studies .................................................................................................33
4.1 Selection of the relevant literature.......................................................................33
4.2 Cohort studies and nested case-control studies reporting on lung cancer ...........34
4.2.1 Overview of the methodologies and study characteristics ......................34
4.2.2 Study quality and utility evaluation ........................................................37
This draft document should not be construed to represent final NTP determination or policy
xi
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
4.2.3 Cancer assessment: Lung ........................................................................38
4.3 Case-control studies ............................................................................................52
4.3.1 Overview of the methodologies and study characteristics ......................52
4.3.2 Study quality and utility evaluation ........................................................52
4.3.3 Cancer assessment: Esophageal cancer ..................................................53
4.4 Cancer assessment: Other cancers .......................................................................59
4.4.1 Other aerodigestive cancers - oral cavity, pharyngeal, and laryngeal cancers
.................................................................................................................59
4.4.2 Other cancers ..........................................................................................60
4.5 Preliminary listing recommendation ...................................................................60
5
Studies of Cancer in Experimental Animals ................................................................63
5.1 Carcinogenicity studies .......................................................................................63
5.1.1 Overview of the studies ..........................................................................63
5.1.2 Study quality assessment ........................................................................65
5.1.3 Assessment of neoplastic findings from carcinogenicity studies ...........68
5.2 Co-carcinogenicity studies ................................................................................101
5.2.1 Overview of the studies ........................................................................101
5.2.2 Overview of the assessment of study quality and utility ......................102
5.2.3 Assessment of findings from co-carcinogenicity studies......................102
5.3 Synthesis of the findings across studies ............................................................103
6
Mechanistic and Other Relevant Effects....................................................................107
6.1 Cobalt particles and cobalt ions.........................................................................107
6.2 Proposed modes of action of cobalt carcinogenicity .........................................114
6.2.1 Genotoxicity, inhibition of DNA repair, and related effects ................116
6.2.2 Oxidative stress .....................................................................................121
6.2.3 Hypoxia mimicry and HIF-1 stabilization ............................................123
6.2.4 Cell signaling and gene expression modulation....................................125
6.3 Cobalt tissue levels from patients with lung and other cancers ........................126
6.4 Synthesis ............................................................................................................126
7
Overall Cancer Evaluation and Preliminary Listing Recommendation .....................127
7.1 Mechanistic and other relevant data ..................................................................127
7.2 Cobalt and certain cobalt compounds as a class ................................................127
7.2.1 Physicochemical properties and toxicokinetics ....................................128
7.2.2 Toxicological effects and key events ....................................................129
7.2.3 Overall synthesis ...................................................................................130
7.3 Evidence of carcinogenicity from studies in experimental animals ..................131
7.4 Evidence of carcinogenicity from studies in humans ........................................132
8
References ..................................................................................................................133
Part 2
Draft Profile .........................................................................................................................1
xii
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
List of Tables
Table 1-1. Physical and chemical properties for cobalt metal and representative cobalt
compoundsa ..................................................................................................................3
Table 2-1. U.S. cobalt compounds production volumes for 2012 exceeding 100,000 pounds per
yeara ..............................................................................................................................6
Table 2-2. U.S. imports and exports of cobalt compounds for 2013 (converted from kg by NTP) 6
Table 2-3. 2012 U.S. consumption and use pattern for cobalt .........................................................7
Table 2-4. Uses for representative inorganic and organic cobalt compounds .................................8
Table 2-5. Workplace air levels of cobalt ......................................................................................12
Table 2-6. Cobalt content of some foods .......................................................................................14
Table 3-1. Interspecies comparison of lung retention of cobalt oxide (Co3O4) .............................26
Table 3-2. Elimination half-lives for cobalt administered to experimental animals......................30
Table 4-1. Cohort and nested case-control studies of exposure to cobalt ......................................36
Table 4-2. Bias and quality summary for cohort and nested case-control studies .........................38
Table 4-3. Evidence from cohort and case-control studies on lung cancer and exposure to cobalt40
Table 4-4. Case-control biomarker studies of exposure to cobalt .................................................52
Table 4-5. Bias and quality summary for case-control studies ......................................................53
Table 4-6. Evidence from studies of aerodigestive cancers and exposure to cobalt......................55
Table 5-1. Overview of cancer studies in experimental animals reviewed ...................................64
Table 5-2. Overview of experimental animal carcinogenicity study quality evaluations..............65
Table 5-3. Lung neoplasms and non-neoplastic lesion in experimental animals exposed to cobalt
compounds..................................................................................................................70
Table 5-4. Injection site neoplasms and non-neoplastic lesions in experimental animals exposed
to cobalt compounds ...................................................................................................83
Table 5-5. Other and distal site neoplasms and relevant non-neoplastic lesions in experimental
animals exposed to cobalt compounds .......................................................................93
Table 5-6. Overview of co-carcinogenicity studies in experimental animals reviewed ..............101
Table 5-7. Overall results of carcinogenicity studies in experimental animals sorted by cobalt
compound .................................................................................................................105
Table 6-1. In vitro mechanistic data comparing effects of cobalt nanoparticles, microparticles,
and ions.....................................................................................................................108
Table 6-2. Summary assessment of genotoxicity and related effects for cobalt compounds ......118
Table 7-1. Comparison of chemical and biological properties of cobalt metal and cobalt
compounds................................................................................................................131
List of Figures
Figure 2-1. Cluster graph of urine cobalt levels ..............................................................................9
Figure 2-2. Cluster graphs of cobalt levels in hair .........................................................................10
Figure 2-3. Flow of cobalt released from anthropogenic processes ..............................................16
Figure 3-1. Cobalt disposition........................................................................................................19
Figure 3-2. Rate of translocation of cobalt from lung to blood following inhalation of cobalt
oxide particles.............................................................................................................25
Figure 4-1. Forest plot showing risk ratios (SIR, SMR or OR as noted) and 95% CI for
epidemiological studies of cobalt exposure................................................................51
Figure 6-1. Proposed modes-of-action of cobalt carcinogenicity. ...............................................115
This draft document should not be construed to represent final NTP determination or policy
xiii
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Figure 6-2. K-ras mutations in lung tumors from cobalt-exposed and non-exposed rodents .....122
xiv
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
1 Chemical identification and properties
The candidate substance being reviewed in this monograph is “Cobalt and Certain Cobalt
Compounds.” “Certain” refers to those cobalt compounds that release cobalt ions in vivo. The
available database on cobalt and cobalt compounds vary by cobalt form; however, there are
carcinogenicity, genotoxicity and toxicity studies on cobalt metal and of some water-soluble and
poorly soluble compounds. Of note, are the two NTP bioassay studies, one with a very soluble
cobalt compound, cobalt sulfate (NTP 1998), and one with cobalt metal (NTP 2014b). Together,
the carcinogenicity, genotoxicity, and other mechanistic information on these representative
forms of cobalt inform the discussion in this document on cobalt and certain cobalt compounds.
Water-soluble cobalt compounds dissolve in the fluids outside cells for cellular uptake, while
particles of poorly soluble cobalt compounds can be taken up intact by cells and release ions
within the cell (see Table 1-1). Of note, vitamin B12, which is an essential cobalt-containing
nutrient, does not meet the criteria for “certain cobalt compounds” because it does not release
cobalt ions in acidic gastric or lysosomal fluids and passes through the body intact while bound
to specific carrier proteins (Neale 1990). Cancer and toxicity studies of some forms of cobalt,
such as cobalt alloys and radioactive forms of cobalt, are excluded from this review because of
confounding exposures.
Cobalt (Co) is a naturally occurring transition element with magnetic properties. It is the 33rd
most abundant element and comprises approximately 0.0025% of the weight of Earth’s crust.
Cobalt is a component of more than 70 naturally occurring minerals including arsenides,
sulfides, and oxides. The only stable and naturally occurring cobalt isotope is 59Co. Metallic
cobalt, Co(0), exists in two allotropic forms, hexagonal and cubic, which are stable at room
temperature (WHO 2006, ATSDR 2004, IARC 1991). Cobalt predominantly occurs in two
oxidation states, +2 (Co(II)) and +3 (Co(III)).
1.1
Properties of cobalt metal and cobalt compounds, both soluble and poorly soluble
Table 1-1 presents physical and chemical properties (molecular weight, crystalline form, density
or specific gravity, water solubility, and bioaccessibility) for cobalt and cobalt compounds for
which animal or genotoxicity testing data are available or that are in commercial use greater than
100,000 pounds per year in the United States (per EPA Chemical Data Reporting rule). The
physical and chemical properties are divided into three groups, including metals, soluble cobalt
compounds, and poorly soluble cobalt compounds, to provide a framework for relating chemicals
for which potential biological effects are unknown to chemicals for which biological effect data
are available.
1.2
Water solubility and bioaccessibility
Evaluation of toxicological and carcinogenic effects of cobalt compounds depends largely on the
release of cobalt ions that can be transported to and taken up at target sites or released within
cells from particles (see Section 6). Cobalt sulfates, chlorides, and nitrates tend to be soluble in
water, while oxides (including the mixed oxide, Co3O4), hydroxides, and sulfides tend to be
poorly soluble or insoluble (Lison 2015). Organic cobalt compounds can be either soluble (e.g.,
cobalt(II) acetate) or insoluble (e.g., cobalt carbonate, cobalt(II) oxalate) (CDI 2006) (see Table
This draft document should not be construed to represent final NTP determination or policy
1
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
1.1). The water solubility of cobalt compounds is largely pH dependent, and cobalt is generally
more mobile in acidic solutions than in alkaline solutions.
Co(0) metal nano- and microparticles dissolve in cell-free culture medium in a concentrationand time-dependent manner while cobalt oxide particles are practically insoluble in water or
culture medium (Ortega et al. 2014, Sabbioni et al. 2014, Ponti et al. 2009). Smaller particles
dissolve faster than larger particles (Lison 2015, Kyono et al. 1992).
While water solubility represents a measure of a compound’s tendency to release ions available
for biological uptake, solubilization of some water-insoluble compounds may be enhanced in
biological fluids at low pH and in the presence of binding proteins (IARC 2006) (see below).
The bioavailability (i.e., extent of systemic absorption) of a metal species is determined by its
solubility in biological fluids (Brock and Stopford 2003, Stopford et al. 2003). For metals like
cobalt, with several species with different valence states having dissimilar solubility
characteristics existing in different compounds, in vivo bioavailability testing can be costprohibitive and time consuming. Therefore, in lieu of in vivo testing, measured solubility of
compounds in artificial fluids (i.e., bioaccessibility) can be used as a surrogate for
bioavailability.
Cobalt metal, and several water-soluble compounds (cobalt sulfate heptahydrate and chloride)
and poorly soluble compounds (cobalt oxide, bis(2-ethyl-hexanoate), carbonate, and
naphthenate) were found to be soluble in biological fluids, suggesting that they release cobalt
ions (see the right-hand column of Table 1-1). These bioaccessibility studies for cobalt
compounds have been performed using synthetic equivalents of gastric and intestinal fluids (for
ingestion exposure); alveolar, interstitial, and lysosomal fluids (for inhalation exposure);
perspiration fluids (for dermal exposure); and synovial fluid (for metal joint prostheses),
identified from exposure scenarios including drawing with soft pastels and manufacturing and
use of alloy materials (Hillwalker and Anderson 2014, Brock and Stopford 2003, Stopford et al.
2003). These studies indicate that species of cobalt compound, particle size, surface area, and pH
of the surrogate fluid can impact cobalt solubility (Stopford et al. 2003). In a study of intra- and
inter-laboratory variability of bioaccessibility testing results for metals and metal compounds
including cobalt powder and cobalt oxide in synthetic gastric, perspiration, lysosomal, and
interstitial fluids, results demonstrated overall satisfactory within-laboratory variability;
however, absolute bioaccessibility results in some biological fluids may vary between different
laboratories (Henderson et al. 2014).
Cobalt(II) ions released into solution can form complexes with organic or inorganic anions with
equilibrium conditions determined by activity of electrons (Eh), activity of hydrogen ions (pH),
and anion presence (Smith and Carson 1981). In general, lower pH generates higher free Co(II)
concentrations in solution, and higher pH gives rise to cobalt-carbonate complex formation
(WHO 2006). The in vivo concentration of free Co(II) ions is relatively low because these
cations precipitate in the presence of physiological concentrations of phosphates and
nonspecifically bind to proteins such as albumin (Lison 2015).
2
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 1-1. Physical and chemical properties for cobalt metal and representative cobalt compounds
a
Name
(+2 valence unless otherwise
indicated)
CAS No.
06/05/15
Formula
Molecular
weight
Physical form
Density or
specific
gravity
Solubility (grams
per 100 cc cold
water)
Bioaccessiblity
(% solubility in
gastric/lysosomal
fluids)
Metal
Cobalt
7440-48-4
Co
58.9
Grey hexagonal or cubic metal
8.92
.000875
100/100
Cobalt nanoparticles
7440-48-4
Co
58.9
–
–
–
–
Sulfate heptahydrate
10026-24-1
CoSO4•7H2O
281.1
Red pink, monocl.
1.95
60.4
100/100
Chloride
7646-79-9
CoCl2
129.9
Blue hexagonal leaflets
3.36
45
100/100
71-48-7
Co(C2H2O2)2
249.1
Red-violet, monocl.
1.70
s
–
10141-05-6
CoN2O6
182.9
Red powder or crystals
2.49
s
–
Oxide
(II, III) Oxide
1307-96-6
1308-06-1
CoO
Co3O4
74.9
240.8
Green-brown cubic
Black, cubic
6.45
6.07
i
i
100/92.4
2-ethyl-hexanoate (org.)
136-52-7
Co(C8H15O2)2
173.7
Blue liquid (12% Co)
1.01
–
100/100
Carbonate (org.)
513-79-1
CoCO3
118.9
Red, trigonal
4.13
i
100/100
Naphthenate (org.)
61789-51-3
Co(C11H7O2)2
401.3
Purple liquid (6% Co)
0.97
–
100/100
Hydroxide
21041-93-0
Co(OH)2
93.0
Rose-red, rhomb
3.60
0.00032
–
Sulfide
1317-42-6
CoS
91.0
Reddish octahedral
5.45
0.00038
–
Oxalate (org.)
814-89-1
CoC2O4
147.0
White or reddish
3.02
i
–
Propionate (org.)
1560-69-6
Co(C3H5O2)2
205.1
–
–
–
–
Stearate (org.)
1002-88-6
Co(C18H35O2)2
625.9
–
–
–
–
Soluble cobalt compounds
Acetate (org.)
Nitrate
Poorly soluble compounds
Sources: SciFinder; PubChem Compounds Database; ChemIDplus Database; Cobalt Development Institute (CDI) Report (2006); Hazardous Substances Data Bank (HSDB);
Stopford et al. (2003).
org. = organic compound; all others are inorganic.
a
Cobalt forms or compounds tested for carcinogenicity or genetic toxicity, or for possible mechanisms of action are italicized.
This draft document should not be construed to represent final NTP determination or policy
3
06/05/15
1.3
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Variability of valence
As noted above, cobalt exists primarily as (Co(II)) and (Co(III)), and Co(II) is much more stable
in aqueous solution (Paustenbach et al. 2013, Nilsson et al. 1985). Electron-donor ligands (e.g.,
NH3) can stabilize Co(III) in aqueous solution (IARC 1991). In acid solution, Co(II) is the stable
form in the absence of electron-donor ligands, and Co(III) is so unstable that it quickly reduces
to Co(II), oxidizing water and liberating oxygen. In contrast, air or hydrogen peroxide can
oxidize Co(II) to the more stable Co(III) complex in alkaline solutions containing ammonium
hydroxide or cyanide. This interconversion between Co(II) and Co(III) is important in the use of
cobalt compounds as catalysts and paint driers, and in biological reactions involving vitamin B12
(Paustenbach et al. 2013, IARC 1991).
Cobalt is present in its stable +2 valence state in the environment and in most commercially
available cobalt compounds, with the exception of the mixed oxide (Co(II,III) or Co3O4)
(Paustenbach et al. 2013, IARC 1991). Some simple salts of cobalt in its +3 valence state (e.g.,
Co2O3) have been used commercially. Cobalt compounds of commercial and toxicological
interest include cobalt metal, alloys, and composite materials; oxides (e.g., cobalt oxide and
tetraoxide); and salts (e.g., cobalt(II) chloride, sulfide, and sulfate) (Lison 2015). Important salts
of carboxylic acids include formate, acetate, citrate, naphthenate, linoleate, oleate, oxalate,
resinate, stearate, succinate, sulfamate, and 2-ethylhexanoate.
Cobalt can also exist in -1, +1, and +4 oxidation states (Nilsson et al. 1985). Cobalt is in its -1
state in cobalt carbonyls such as [Co(CO)4]H and in cobalt-nitrosyls, in its +1 state in some
cobalt-cyanide complexes, and in its +4 state in compounds with cobalt bonded to fluoride or
oxygen.
1.4
Summary
Both cobalt metal, as particles or nanoparticles, have been found to be 100% bioaccessible (i.e.,
dissolving to release cobalt ions) in artificial gastric and lysosomal fluids. The soluble
compounds, cobalt(II) sulfate heptahydrate and cobalt(II) chloride, and the poorly soluble
compounds, cobalt(II) oxide, cobalt bis(2-ethyl hexanoate), cobalt carbonate, and cobalt
naphthenate, also were completely (or almost completely) soluble in the two acidic fluids. The
metals and poorly soluble compounds tended to be less bioaccessible in neutral biological fluids,
which is consistent with the pH dependence for releasing cobalt ions in solution.
4
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
2 Human Exposure
This section describes cobalt mining and production (Section 2.1); use (Section 2.2);
biomonitoring and exposure to cobalt and cobalt compounds (Section 2.3); exposure in the
workplace (Section 2.4); potential exposure from other sources such as food, consumer products,
tobacco, and medical products (Section 2.5); and potential for environmental exposure (Section
2.6). The material presented in Sections 2.1 through 2.6 is summarized in Section 2.7. Although
studies of cobalt alloys were not considered informative for either animal tumor studies or
human carcinogenicity studies because they are not specific for exposure only to cobalt, humans
can be exposed to cobalt from producing or using these compounds, and thus those exposures are
discussed in this section.
2.1
Mining and production
Cobalt is most often found in ores associated with copper or nickel, but may also be a by-product
of zinc, lead, and platinum-group metals (CDI 2006, Davis 2000). Cobalt-containing ores often
contain arsenic, such as safflorite, CoAs2; skutterudite, CoAs3; erythrite, Co3(AsO4)2•8H2O; and
glaucodot, CoAsS (CDI 2006, ATSDR 2004, Davis 2000). The largest cobalt reserves are in the
Congo (Kinshasa), Australia, Cuba, Zambia, Canada, Russia, and New Caledonia (Shedd 2014a).
Most U.S. cobalt deposits are in Minnesota, but other important deposits are in Alaska,
California, Idaho, Missouri, Montana, and Oregon. Except for Idaho and Missouri, future
production from these deposits would be as a by-product of another metal.
Except for a negligible amount of by-product cobalt produced as an intermediate product from
mining and refining platinum-group metals ore, the United States did not refine cobalt in 2012
(Shedd 2014b). Since 2009, no cobalt has been sold from the National Defense Stockpile. In
2012, 2,160 metric tons of cobalt was recycled from scrap. Cobalt has not been mined in the
United States in over 30 years (ATSDR 2004); however, a primary cobalt mine, mill, and
refinery are currently being established in Idaho that will produce more than 1,500 tons of highpurity cobalt metal annually to capitalize on increasing cobalt demand driven in part by growth
in “green” energy technology (e.g., rechargeable batteries for electric and hybrid electric vehicles
or portable electronics applications (Farquharson 2015, Mining Technology Market and
Customer Insight 2015, Rufe 2010) Based on a presentation dated May 2015, preliminary work
on the site has been completed (Formation Metals Inc. 2015).
Cobalt and several cobalt compounds are high-production-volume chemicals based on their
production or importation into the United States in quantities of 1 million pounds or more per
year. Table 2-1 shows U.S. cobalt and cobalt compound production volumes for 2012 that
exceed 100,000 pounds per year; the highest United States production volume is for cobalt
(7440-48-4) (23,384,002 lb). Table 2-2 lists recent U.S. imports and exports of cobalt and cobalt
compounds; the highest import value is for “unwrought cobalt excluding alloys, including
powders” (16,151,599 lb) and the highest export value is for “cobalt, wrought, and articles
thereof” (4,841,750 lb).
This draft document should not be construed to represent final NTP determination or policy
5
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
a
Table 2-1. U.S. cobalt compounds production volumes for 2012 exceeding 100,000 pounds per year
CAS Numberb
Quantity (lb)c
Cobalt compound
7440-48-4
Cobalt
23,384,002
21041-93-0
Cobalt hydroxide (Co(OH)2)
4,709,137
136-52-7
Cobalt 2-ethylhexanoate
4,294,523
1307-96-6
Cobalt oxide (CoO)
1,385,848
513-79-1
Cobalt carbonate
1,038,821
10124-43-3
Cobalt sulfate
1,000,000–10,000,000
10141-05-6
Cobalt nitrate
1,000,000–10,000,000
1308-06-1
Cobalt oxide (Co3O4)
1,000,000–10,000,000
1560-69-6
Cobalt propionate
1,000,000–10,000,000
71-48-7
Cobalt acetate
1,000,000–10,000,000
814-89-1
Cobalt oxalate
600,000
1317-42-6
Cobalt sulfide (CoS)
254,733
61789-52-4
Cobalt tallate
192,900
61789-51-3
Cobalt naphthenate
100,000–500,000
a
Three cobalt compounds for which properties are reported in Table 1-1 are not listed in Table 2-1 because of the production
level or lack of reported production data. Cobalt oxide (11104-61-3) production levels were 94,139 lb in 2012. Cobalt sulfide
(12013-10-4, CoS2) and cobalt chloride (7646-79-9, CoCl2) production levels for 2012 were withheld by the manufacturers.
b
CAS# were identified from multiple sources: ChemIDplus Database; EPA Chemical Data Reporting (2012); PubChem
Compounds Database; Ullmann's Encyclopedia of Industrial Chemistry (2012).
c
EPA Chemical Data Reporting (2012). See reference list for specifics.
Table 2-2. U.S. imports and exports of cobalt compounds for 2013 (converted from kg by NTP)
Cobalt-compound/category
U.S. imports (lb)
U.S. exports (lb)
342,918
520,996
1,193,856
–a
Cobalt chloride
215,661
14,304
Cobalt ores and concentrates
82,376
1,004,825
Cobalt oxides and hydroxides; commercial cobalt oxides
5,300,984
902,467
Cobalt sulfate
1,319,004
–a
Cobalt waste and scrap
1,549,151
1,557,515
550,887
4,841,750
Other cobalt mattes and intermediate products of cobalt
metallurgy; powders
1,992,434
–a
Unwrought cobalt alloys
2,132,331
–a
Unwrought cobalt excluding alloys, including powders
16,151,599
–a
Cobalt acetates
Cobalt carbonates
Cobalt, wrought, and articles thereof
Source: (USITC 2014).
a
No specific Schedule B code identified.
2.2
Use
Cobalt is used in numerous commercial, industrial, and military applications. On a global basis,
the largest use of cobalt is in rechargeable battery electrodes; however, rechargeable battery
6
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
production in the United States has been very limited (NIST 2005). In 2012, the reported U.S.
consumption of cobalt was approximately 8,420 metric tons (Shedd 2014b) for the uses shown
below in Table 2-3.
Table 2-3. 2012 U.S. consumption and use pattern for cobalt
Consumption (metric tons
cobalt content)
Percent of total consumption (%)
Superalloys
4,040
48
Chemical and ceramic
2,300
27.3
774
9.2
Other alloys
699
8.3
Steels
548
6.5
Miscellaneous and unspecified
63
0.7
End use
Cemented carbides
a
Source: (Shedd 2014b).
a
Includes magnetic, nonferrous, and wear-resistant alloys and welding materials.
The main uses of cobalt can be grouped into the following general categories: metallurgical;
cemented carbides and bonded diamonds; chemicals; and electronics and “green” energy (CDI
2006). Cobalt nanoparticles are used for medical applications (e.g., sensors, MRI contrast
enhancement, drug delivery), and nanofibers and nanowires also are being used for industrial
applications.
Uses for cobalt compounds with their reported levels of U.S. production are listed in Table 2-1
above. Table 2-4 lists uses for several cobalt compounds for which no production information
was identified.
2.2.1
Metallurgical uses
Metallurgical uses of cobalt include use in superalloys; magnetic alloys, low expansion alloys,
nonferrous alloys, steels, coatings, and bone and dental prostheses (IARC 1991, Davis 2000, CDI
2006, Ohno 2010). Support structures for heart valves are also manufactured from cobalt alloys
(IARC 1991).
2.2.2
Cemented carbides and bonded diamonds
Cemented tungsten carbides (“hard metals”) are composites of tungsten carbide particles (either
tungsten carbide alone or in combination with smaller amounts of other carbides) with metallic
cobalt powder as a binder, pressed into a compact, solid form at high temperatures by a process
called sintering (NTP 2009, IARC 1991). Cobalt is also used in diamond tools from steel with
microdiamonds impregnated into a surface cobalt layer (CDI 2006, IARC 2006).
2.2.3
Chemical uses
Uses of cobalt compounds include as pigments for glass, ceramics, and enamels, as driers for
paints, varnishes, or lacquers, as catalysts, as adhesives and enamel frits (naphthenate, stearate,
oxide), as trace mineral additives for animal diets, and in rechargeable batteries (see Section
2.2.4) (CDI 2006, WHO 2006, ATSDR 2004, IARC 1991) (see Table 2-4). Compounds of
commercial importance are the oxides, hydroxide, chloride, sulfate, nitrate, phosphate,
This draft document should not be construed to represent final NTP determination or policy
7
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
carbonate, acetate, oxalate, and other carboxylic acid derivatives (IARC 1991). A past use of
cobalt (as cobalt sulfate) was as an additive in some beers to increase the stability of the foam
(NTP 1998).
Table 2-4. Uses for representative inorganic and organic cobalt compounds
X
Batteries
X
Driers
X
X
X
X
X
Pigments
X
X
X
X
X
Propionate
2-ethyl-hexanoate
X
X
X
X
X
X
X
Carbonate
Animal diets
Catalysts
Sulfate
X
Acetate
Adhesives
Organic
Oxides
Nitrate
Use
Hydroxide
Chloride
Inorganic
X
X
X
X
X
X
Sources: CDI 2006, Donaldson and Beyersmann 2012, Richardson and Meshri 2001.
2.2.4
Electronics and “green” energy
Due to increased demand for portable rechargeable electronic devices, one of the fastest growth
areas for cobalt use is in high-capacity, rechargeable batteries (Shedd 2014b, CDI 2006, Davis
2000). Cobalt is used in nickel-cadmium (Ni-Cd), nickel-metal hydride (Ni-MH), and lithiumion (Li-ion) battery technologies. Applications for batteries containing cobalt compounds include
portable computers, mobile telephones, camcorders, toys, power tools, and electric vehicles.
Cobalt is also used in integrated circuit contacts and leads and in the production of
semiconductors (CDI 2006, IARC 1991).
Cobalt is the key element in several forms of “green” energy technology applications including
gas-to liquid (GTL) and oil desulfurization (see Section 2.2.2), coal-to liquid (CTL), clean coal,
solar panels, wind and gas turbines, and fuel cells (Rufe 2010). Research is ongoing on use of
cobalt-based catalysts in sunlight-driven water splitting to convert solar energy into electrical and
chemical energy (Deng and Tüysüz 2014).
2.3
Biomonitoring for cobalt
Evidence for widespread exposure to cobalt and cobalt compounds comes from biological
monitoring data measuring cobalt levels in urine, blood, hair, nails, and tissues in individuals
exposed to cobalt from occupational and non-occupational sources (see Appendix B, Tables B-1
and B-2 for levels reported in these studies, source of exposure, and geographical location, and
Figures 2.1 and 2.2). Several of the studies are of people residing in the United States, and thus
demonstrate U.S. exposure.
2.3.1
Evidence of exposure
Urine
8
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Studies measuring cobalt in the urine of people exposed to cobalt from different sources indicate
that the highest levels were generally seen for occupational exposures and failed hip implants;
with lower levels from exposure from normal implants, the environment, or in the general public
(source of exposure unknown). (See Figure 2.1, which depicts the mean (or median) levels of
urinary cobalt in these populations from the studies reported in Appendix B, Table B-1.) The
geometric mean urinary cobalt concentration for the U.S. general public for the most recent
(2011-2012) National Health and Nutrition Examination Survey (NHANES) year for which data
are available is 0.326 µg/L; urinary cobalt measurements in the U.S. general public have
remained consistent since 1999, with a geometric mean value of 0.316 to 0.379 µg/L (CDC
2015).
Figure 2-1. Cluster graph of urine cobalt levels
Filled symbols = U.S. data; open symbols = non-U.S. data.
Hair
Reported mean levels of cobalt in hair are highest among workers and among patients with
unstable hip implants (Figure 2-2). Cobalt levels in samples from patients with stable hip
implants are next highest, with levels taken from populations at risk of environmental exposure
and in the general public being the lowest. Measurements of cobalt in hair in the latter groups
overlap significantly; while one study indicates that cobalt levels among environmentally
exposed populations are similar to levels in workers.
This draft document should not be construed to represent final NTP determination or policy
9
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Figure 2-2. Cluster graphs of cobalt levels in hair
Filled symbols = U.S. data; open symbols = non-U.S. data.
2.3.2
Methods
Urine
Urinary cobalt is considered a good indicator of absorbed cobalt (IARC 2006, WHO 2006),
especially from recent exposures (ATSDR 2004). Urinary cobalt levels are more reflective of
recent exposure for soluble compounds than less soluble compounds (ATSDR 2004).
Blood
Blood cobalt levels are also more reflective of recent exposure for soluble compounds than less
soluble compounds (ATSDR 2004). Although measurements of cobalt in whole blood, plasma,
and serum have been reported by investigators, no consensus seems to exist for which of these
provides the best relationship with levels of exposure to cobalt.
Hair
Although cobalt concentrations in hair provide a potential biomarker for cobalt exposure over
time, the source of exposure in the studies is not known. Because hair fixes trace elements in a
permanent, chemically homogeneous matrix, hair samples reflect a time-integrated exposure
(i.e., current and past exposure levels) over the previous few months, depending on the length of
the hair sample (Suzuki and Yamamoto 1982) and hair metal contents provides a better estimate
than blood in assessing the environmental risk to toxic metals for infrequent and highly variable
exposures (Bax 1981, Petering et al. 1973). The average concentration of cobalt in hair is over
100 times greater than that in blood (Underwood 1977). Average metal concentration can be
obtained by measuring bulk concentration from a length of hair equal to a few weeks’ growth, by
measuring the variation along the length of long hair equal to several months (Suzuki and
Yamamoto 1982), or by taking periodic samples over time (Laker 1982).
10
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Nails
Toenail clippings reflect time-integrated exposure occurring in the timeframe of 12 to 24 months
prior to clipping, and thus are useful biomarkers of exposure when a single sample is assumed to
represent long-term exposure (He 2011, Fleckman 1985). However, toenails generally provide
larger samples and represent more distant past exposures because they take longer to grow out.
Nails are considered to be relatively sheltered from environmental contaminants (relative to hair,
which, though formed from the same keratinous tissue of nail, can be contaminated by dyeing,
bleaching, and permanent waving). Toenails are also more convenient to collect and store than
blood (Garland et al. 1993). However, nails can become contaminated through the use of nail
polishes, some medications, and use of contaminated cutters to produce clippings (He 2011).
Tissue
Several publications measure trace metals (e.g., heavy metals and essential metals) in tissue from
cancer patients with a referent group or tissue. Several clinical surveys have compared levels of
cobalt in cancer patients and non-cancer patients (see Appendix B, Table B-3).
Analytical methods
Analytical methods for cobalt in biological materials include graphite furnace atomic absorption
spectrometry (GF-AAS), inductively coupled plasma-atomic emisson spectrometry (ICP-AES),
differential pulse cathodic stripping voltammetry (DPCSV), and colorimetric determination
(ATSDR 2004). The colorimetric method generally has limited utility because it has poor
sensitivity (Alessio and Dell’Orto 1988). The ICP-AES method is used by NIOSH for exposure
to elements in blood and urine (NIOSH 1994), and NHANES uses a related method of
inductively coupled plasma-mass spectrometry (ICP-MS) for urine heavy metals. With the
exception of the colorimetric method, these methods require wet (acid) digestion followed by
flame ionization to liberate free cobalt ions for detection of total cobalt. Thus, in any biological
sample, the original form of the cobalt, whether inorganic cobalt or part of an organic molecule
like vitamin B12, cannot be determined with these methods (IARC 2006, WHO 2006).
2.4
Characterization of exposure in the workplace
The primary route of occupational exposure to cobalt is via inhalation of dust, fumes, or mists or
gaseous cobalt carbonyl; however, dermal contact with hard metals and cobalt salts can result in
systemic uptake. Occupational exposure to cobalt occurs during the refining of cobalt; during the
production of cobalt powder; during the production and use of cobalt alloys; during hard metal
production, processing, and use; during the maintenance and re-sharpening of hard metal tools
and blades; during the manufacture and use of cobalt-containing diamond tools; and during the
use of cobalt-containing pigments and driers. Workers regenerating spent catalysts may also be
exposed to cobalt sulfides.
Occupational exposure has been documented by measurements of cobalt in ambient workplace
air, in worker blood and urine, and in deceased worker lung tissue (CDC 2013, IARC 2006,
ATSDR 2004, IARC 1991). The NIOSH National Occupational Exposure Survey (NOES)
estimated that approximately 386,500 workers were potentially exposed to cobalt and cobalt
compounds (NIOSH 1990). The survey was conducted from 1981 to 1983, and the NOES
This draft document should not be construed to represent final NTP determination or policy
11
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
database was last updated in July 1990. Table 2-5 reports workplace air levels of cobalt for
various industrial uses of cobalt or cobalt compounds.
Several exposure scenarios arise from the use of cobalt and cobalt compounds in numerous
commercial, industrial, and military applications. The scenarios that generally contribute most to
U.S. releases of cobalt and cobalt compounds as reported to EPA (TRI 2014d) include gold,
copper, and nickel ore mining, hazardous waste treatment and disposal, non-ferrous metal
smelting and refining, fossil fuel electric power generation, and chemical operations (e.g.,
petrochemical manufacturing and synthetic dye and pigment manufacturing). Other potential
exposure scenarios (e.g., copper smelting) exist, but no air data were identified.
Table 2-5. Workplace air levels of cobalt
Cobalt in workplace air mean
(range) in µg/m3
Reference
Use of cobalt-containing diamond
tools (NR)
(0.1–45)
(van den Oever et al. 1990)
Use of cobalt-containing diamond
tools (Italy)
690
115 (with improved ventilation)
(Ferdenzi et al. 1994)
Production of Stellite, a cobaltcontaining alloy (NR)
Several hundred µg/m3
(Simcox et al. 2000)
Production of Stellite, a cobaltcontaining alloy (NR)
9
(Kennedy et al. 1995)
Welding with Stellite, a cobaltcontaining alloy (NR)
160
(Ferri et al. 1994)
Painting porcelain plates with
cobalt compounds (Denmark)
80
26 (after Danish surveillance
program)
(Christensen 1995, Poulsen et al.
1995, Christensen and Poulsen
1994)
Production of cobalt metal and
cobalt salts (Belgium)
127.5 (2–7,700)
(Swennen et al. 1993)
Recycling batteries to recover
cobalt (NR)
Up to 10
(Hengstler et al. 2003)
Production of cobalt salts
(Russian Federation)
(0.05–50)
(Talakin et al. 1991)
Nickel refining (Russian
Federation)
Up to 4
(Thomassen et al. 1999)
Production of cobalt metal and
cobalt salts (Finland)
< 100
(Linna et al. 2003)
Conversion of cobalt metal to
cobalt oxide (South Africa)
9,900 (highest reported)
(Coombs 1996)
Nickel refining (Norway)
< 150a
(Grimsrud et al. 2005)
Exposure scenario (Country)
Source: (IARC 2006).
NR = Not reported.
a
Reported as 0.15 mg/m3. Among the 3,500 personal samples from the breathing zone taken, cobalt values above 50 mg/m3
[50,000 µg/m3] (3 measurements) were excluded.
12
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
2.4.1
06/05/15
Hard-metal production, processing, and use
Air levels of cobalt vary across different stages of the hard metals manufacturing process, with
levels for operations involving cobalt metal powder often reaching maximum levels between
1,000 and 10,000 µg/m3 (NTP 2009). One extreme value of 438,000 µg/m3 was reported for
weighing and mixing operations in a plant in the United States (Sprince et al. 1984). Continuous
recycling of coolants used during the grinding of hard-metal tools after sintering and during
maintenance and re-sharpening has been reported to increase concentrations of dissolved cobalt
in the metal-working fluid, which can be a source of exposure to ionic cobalt in aerosols from the
coolants (IARC 2006). Wet grinding processes are reported to produce higher cobalt
concentrations than dry grinding processes due to coolant mist emissions.
2.4.2
Cobalt-containing diamond tools and cobalt alloys
Diamond polishers inhale metallic cobalt, iron, and silica from the use of cobalt discs to polish
diamond jewels. Cobalt concentrations in workplace air have been reported to range from 0.1 to
45 µg/m3 in diamond jewel polishing and as high as 690 µg/m3 in wood and stone cutting (air
concentrations dropped to 115 µg/m3 after implementation of ventilation system improvements
in the wood and stone cutting factory) (IARC 2006).
Occupational exposure results from production and use (e.g., welding, grinding, and sharpening)
of cobalt alloys. Concentrations of cobalt in workplace air of facilities producing and using
Stellite have been reported to range from 9 to several hundred micrograms per cubic meter
(IARC 2006).
2.4.3
Pigments
Cobalt concentrations in workplace air at Danish porcelain factories using cobalt-aluminate
spinel or cobalt silicate dyes have been reported to exceed the Danish hygienic standard by 1.3to 172-fold (Tüchsen et al. 1996) (see Section 4). Due to improvements made to workplace
conditions in the 1982 to 1992 time period, concentrations of cobalt in workplace air decreased
from 1,356 nmol/m3 [80 µg/m3] to 454 nmol/m3 [26 µg/m3] and worker urinary cobalt decreased
from 100-fold to 10-fold above median concentration of controls (IARC 2006, 1991).
2.4.4
Cobalt production (metals and salts)
Cobalt concentrations in workplace air have been reported to range from 2 to 50,000 µg/m3 from
hydrometallurgical purification, battery recycling, and cobalt compound (acetate, chloride,
nitrate, oxide, and sulfate) production. Worker urinary cobalt for these facilities ranged from 1.6
to 2,038 µg/g creatinine (IARC 2006).
2.5
Surgical implants
Surgical implantation (e.g., orthopedic joint replacements) can result in exposure to cobalt. The
total number of hip replacements in the United States has been variously reported as 120,000 per
year (Polyzois et al. 2012) or 400,000 per year (Devlin et al. 2013, Frank 2012). Blood cobalt
ion concentrations generally increase by 5- to 10-fold from preoperative to postoperative levels
(Polyzois et al. 2012) (see Table B-1 for cobalt levels in blood from individuals with implants).
Metal-on-metal (MoM) implants are reportedly associated with blood cobalt less than 4 µg/L and
urine levels less than 7.5 µg/L (Jantzen et al. 2013, Sampson and Hart 2012).
This draft document should not be construed to represent final NTP determination or policy
13
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
One in eight total hip implants requires revision within 10 years, and 60% of those are due to
wear-related complications (Bradberry et al. 2014). All hip implants that contain metal
components contain cobalt as part of cobalt-chromium-molybdenum alloys (Devlin et al. 2013,
Sampson and Hart 2012). Release of metal from implants results from both wear and corrosion,
which is caused by body fluids contacting the metal surfaces or by formation of an
electrochemical couple between different metal components (Sampson and Hart 2012). Implants
that have failed because of excessive wear or corrosion have been associated with systemic
cobalt toxicity in some cases, and cobalt levels in some of these individuals have been reported.
Blood levels associated with toxicity may be related to the type of implant; levels were higher
among 10 patients with failed ceramic prosthesis (median blood concentration = 506 µg/L; range
= 353 to 6,521) compared with eight patients with toxicity from implanted MoM devices
(median 34.5 µg/L range = 13.6 to 398.6) (Bradberry et al. (2014)) Peak blood cobalt
concentrations were > 250 µg/L. The Medicines and Healthcare Products Regulatory Agency
(MHRA) in the United Kingdom issued a safety alert that proposed a level of 7 µg/L cobalt as an
action level for further clinical investigation and action (MHRA 2012) and at 10 µg/L in the
United States by the Mayo Clinic (Mayo Clinic 2015).
2.6
Other sources of exposure: Food, consumer and other medical products and tobacco
Exposure to cobalt in the general population also occurs via inhalation of ambient air and
ingestion of drinking water; however, food is the largest source of cobalt exposure in the general
population (ATSDR 2004). Daily cobalt intake from food has been reported to range from 5 to
50 µg/day (Lison 2015). Green, leafy vegetables and fresh cereals contain the most cobalt, and
dairy products, refined cereals, and sugar contain the least (IARC 1991) (see Table 2-6). Cobalt
compounds were added to beer in the past, but this use has been discontinued. Reported values
for cobalt content of foods can vary due to differences in environmental cobalt levels, analytical
difficulties, and inadequate analytical techniques.
Table 2-6. Cobalt content of some foods
Cobalt content
Food
(micrograms per 100 grams of food)
Green, leafy vegetables
20–60
Organ meats
15–25
Muscle meats
7–12
Dairy products, refined cereals, sugar
1–3
Sources: (Briggs and Wahlqvist 1998, IARC 1991).
Higher cobalt intake may result from consumption of over-the-counter or prescription vitamin
and mineral preparations (e.g., cobalt chloride or vitamin B12 formulations). In the 1970s, oral
intake of cobalt chloride was used to increase red blood cell counts in anemic patients (but
discontinued when enlarged thyroids and goiters were observed at higher doses). In the last
decade, oral administration of cobalt chloride has been used to correct excessive estrogen
production during female hormone replacement therapy (Tvermoes et al. 2013, Unice et al.
2012, Lippi et al. 2005).
14
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Cobalt is present in consumer products including cleaners, detergents, and soaps (ATSDR 2004).
The NLM Household Products Database listed 6 products containing cobalt as an ingredient: 1
nickel metal hydride battery (5% to 10% cobalt), 4 dishwasher detergents (2 powders and 2
semi-solid pouches containing powder), and 1 spray car wax product (HPD 2014).
Different brands of tobacco have been reported to contain cobalt ranging from < 0.3 to 2.3 µg/g
dry weight; 0.5% of the cobalt content is transferred to mainstream smoke (WHO 2006).
Smokers with no occupational exposure have been reported to have a significantly higher mean
urinary cobalt concentration (0.6 µg/L, SD = 0.6) than non-smokers (0.3 µg/L, SD = 0.1); cobalt
concentrations in blood were the same (Alexandersson 1988, as cited in IARC 1991). However,
examination of urinary cobalt levels between cigarette smoke-exposed and unexposed NHANES
participants for survey years 1999 to 2004 indicates that there was no significant difference in
urinary cobalt levels for smokers and non-smokers (unadjusted for creatinine) (Richter et al.
2009). Richter et al. noted that while cobalt deficiencies were not reported, smoking does
interfere with vitamin B12 absorption.
2.7
Potential for environmental exposure
Information on potential for environmental exposure discussed below includes data for releases
(Section 2.1.7), occurrence (Section 2.7.2), and exposure (Section 2.7.3).
2.7.1
Releases
Approximately 75,000 metric tons of cobalt enters the global environment annually (CDI 2006,
Shedd 1993). Cobalt is released through the natural processes of rock weathering and biological
extraction (i.e., biochemical processes of bacteria and other microorganisms that extract cobalt
from rocks and soils). Figure 2-3 shows cobalt released from anthropogenic processes (i.e.,
burning of fossil fuels, metal production and use). Similar amounts come from natural (40,000
metric tons) and anthropogenic (35,000 metric tons) sources; the majority of the natural source
contribution is from biochemical processes and the majority of the anthropogenic contribution is
from metal production and use.
This draft document should not be construed to represent final NTP determination or policy
15
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Figure 2-3. Flow of cobalt released from anthropogenic processes
Adapted from (CDI 2006, Shedd 1993).
Cobalt’s widespread use in numerous commercial, industrial (e.g., mining and extraction from
ores), and military applications results in releases to the environment through various waste
streams. According to the TRI, total reported on- and offsite release of cobalt and cobalt
compounds was approximately 5.5 million pounds from 723 facilities in 2013 (TRI 2014c,
2014b, 2014a). Calculations based on media-specific release data from TRI indicate that releases
to land accounted for 82% of total releases, offsite disposal for 15%, and underground injection,
air, and water for 1% each in 2013.
2.7.2
Occurrence
The average concentration of cobalt in ambient air in the United States has been reported to be
approximately 0.4 ng/m3 (ATSDR 2004). Levels can be orders of magnitude higher near source
areas (e.g., near facilities processing cobalt-containing alloys, compounds, etc.). Sources of
cobalt in the atmosphere can be natural (e.g., wind-blown continental dust, seawater spray,
volcanoes, forest fires, and marine biogenic emissions), and anthropogenic (e.g., burning of
fossil fuels, mining and smelting of cobalt-containing ores, hazardous waste treatment and
disposal, etc.) (TRI 2014a, EPA 2012, ATSDR 2004).
16
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Median cobalt concentration in U.S. drinking water has been reported to be < 2.0 µg/L; however,
levels as high as 107 µg/L have been reported. It is unclear whether higher levels could indicate
cobalt being picked up in distribution systems (ATSDR 2004). Cobalt concentrations have been
reported to range from 0.01 to 4 µg/L in seawater and from 0.1 to 10 µg/L in freshwater and
groundwater (IARC 2006).
Studies have reported cobalt soil concentrations ranging from 0.1 to 50 ppm. However, soils near
ore deposits, phosphate rock, ore smelting facilities, soils contaminated by airport or highway
traffic, or other source areas may contain higher concentrations (e.g., soil cobalt concentrations
as high as 12,700 ppm reported near hard-metal facilities) (IARC 2006). The soil concentration
of cobalt available to be taken up by plants has been reported to range from 0.1 to 2 ppm (IARC
2006).
2.7.3
Exposure
Information on exposures to cobalt from environmental releases is limited, and no data for U.S.
exposures were identified. Biomonitoring research has confirmed general public exposure to
cobalt in scenarios including non-ferrous metal mining (see Figure 2-1). A study of metal
exposure from mining and processing of non-ferrous metals in Katanga, Democratic Republic of
Congo found that geometric mean urinary cobalt concentrations were 4.5-fold higher for adults
and 6.6-fold higher for children in urban and rural communities near mines and metal smelters
than in rural communities without mining or industrial activities (Cheyns et al. 2014).
2.8
Summary and synthesis
Several lines of evidence indicate that a significant number of people living in the United States
are exposed to cobalt and cobalt compounds. This evidence includes cobalt and several cobalt
compounds being high-production-volume chemicals, widespread use in numerous commercial,
industrial, and military applications, and biological monitoring data (i.e., urine, blood, hair, and
nails) demonstrating exposure in occupationally and non-occupationally exposed populations.
TRI data indicate that production- and use-related releases of cobalt and cobalt compounds have
occurred at numerous industrial facilities in the United States.
Biomonitoring studies measuring cobalt in the urine of people exposed to cobalt from different
sources indicate that the highest levels were generally seen for occupational exposures and
unstable hip implants; lower cobalt levels were due to exposure from stable hip implants or the
environment, or in the general public (source of exposure unknown). In general, levels of cobalt
in blood (including whole blood, plasma, and serum), in hair, and in nails show a similar pattern
to those for urinary cobalt levels.
The primary route of occupational exposure to cobalt is via inhalation of dust, fumes, mists
containing cobalt, or of gaseous cobalt carbonyl. Dermal contact with hard metal and cobalt salts
can result in systemic uptake. Occupational exposure to cobalt occurs during refining of cobalt;
production of cobalt powder; the production and use of cobalt alloys; hard-metal production,
processing, and use; maintenance and re-sharpening of hard-metal tools and blades; manufacture
and use of cobalt-containing diamond tools; and use of cobalt-containing pigments and driers.
Occupational exposure has been documented by measurements of cobalt in ambient workplace
air, worker blood and urine, and deceased worker lung tissue.
This draft document should not be construed to represent final NTP determination or policy
17
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Some of the highest levels of cobalt reported in blood or urine have been associated with failed
medical devices (such as metallic hip implants containing cobalt alloys). Levels of cobalt
reported in blood or urine from stable hip implants are generally less than those reported for
unstable hip implants and occupational exposures but more than those reported for exposures
from the environment or in the general public.
Although exposure to cobalt in the general public can occur via inhalation of ambient air and
ingestion of drinking water, for the majority of the general public the primary source of cobalt
exposure is food; daily cobalt intake from food has been reported to range from 5 to 50 µg/day.
Higher cobalt intake may result from consumption of over-the-counter or prescription mineral
preparations. Other sources of exposure to cobalt and cobalt compounds include some household
consumer products, primarily dishwasher detergents and nickel metal hydride batteries.
18
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
3 Disposition and Toxicokinetics
Disposition and toxicokinetics refer to how a chemical can enter and leave the body, what
happens to it once it is in the body, and the rates of these processes. Section 3.1 discusses the
disposition of cobalt and cobalt compounds in humans and experimental animals, and
toxicokinetic data are presented in Section 3.2. Disposition and toxicokinetic data are important
because they describe various factors that affect the toxicity of a chemical. These factors include
routes and rates of absorption, distribution, and retention; routes of elimination; and gender
and/or species differences in these factors. The mechanistic implications of these data are
discussed in Section 7.
3.1
Disposition
Disposition includes absorption, deposition, distribution, metabolism, retention, and excretion.
The disposition of cobalt is affected by several factors including the chemical form, solubility,
dose, particle size, route of exposure, nutritional status, and age of the species exposed. The
primary exposure, distribution, and excretion pathways of cobalt are illustrated in Figure 3-1.
Data derived from studies in humans are discussed in Section 3.1.1 while studies in experimental
animals are discussed in Section 3.1.2.
Figure 3-1. Cobalt disposition
Source: Adapted from (Keegan et al. 2008)
This draft document should not be construed to represent final NTP determination or policy
19
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Humans
The normal dietary intake of cobalt ranges from about 5 to 50 µg/day, most of which is inorganic
with a very small fraction from vitamin B12 (Lison 2015); the normal range of cobalt
concentrations (nonoccupational exposure) in the blood and urine are about 0.1 to 0.5 µg/L and <
2 µg/L, respectively (Paustenbach et al. 2013, IARC 2006) (see Section 2). About 90% to 95%
of cobalt in blood is bound to serum albumin while the concentration of free cobalt is about 5%
to 12% of the total cobalt concentration (Paustenbach et al. 2013, Simonsen et al. 2012).
Letourneau et al. (1972) showed that a dose of vitamin B12 had no impact on retention of
inorganic cobalt in humans. The total body burden of cobalt in humans is estimated as 1.1 to 1.5
mg with about 85% present in the vitamin B12 organometallic complex (Paustenbach et al. 2013,
WHO 2006).
Absorption
Cobalt absorption from the gastrointestinal (GI) tract is highly variable, with reported values
ranging from < 5% to 97% (Holstein et al. 2015, NTP 2014b, Paustenbach et al. 2013, IARC
2006, WHO 2006, Smith et al. 1972). Unice et al. (2012) suggested a central tendency value of
25% for GI absorption of soluble inorganic cobalt while Unice et al. (2014) assumed GI
absorption of 20% to 45% for aqueous forms and 10% to 25% for solid forms. Cobalt
concentrations in whole blood increased 9 to 36 times above normal background concentrations
in volunteers that ingested a liquid dietary supplement that contained cobalt chloride for up to 16
days (Tvermoes et al. 2013). Soluble cobalt compounds are better absorbed than insoluble forms
(Christensen and Poulsen 1994, Christensen et al. 1993a). For example, men and women
volunteers who ingested tablets containing soluble cobalt chloride (CoCl2) had approximately
10-fold higher concentrations of cobalt in blood and 50- to 90-fold higher concentrations in urine
than when they ingested tablets containing insoluble cobalt oxide (Co3O4) (Christensen et al.
1993a). Controlled studies in human volunteers also indicate that GI uptake is higher in women
than in men with adjusted mean whole blood concentrations about two-fold higher in women
(Finley et al. 2013, Christensen et al. 1993a). The higher cobalt uptake in women may be due to
a higher incidence of iron deficiency since cobalt absorption efficiency is higher in individuals
with iron deficiency (31% to 71% compared to 18% to 44% in control subjects) (Sorbie et al.
1971, Valberg et al. 1969). Meltzer et al. (2010) reported that cobalt whole blood concentrations
were significantly elevated in women with low serum ferritin concentrations compared to women
with higher serum ferritin concentrations and in women with mild to moderate anemia compared
to women with only slightly reduced hemoglobin. Low iron status was a prerequisite for high
blood concentrations of cobalt; however, not everyone with low iron status had increased blood
levels of cobalt. These data suggest that cobalt and iron may share a common gastrointestinal
uptake mechanism that may be upregulated with anemia or iron deficiency (Paustenbach et al.
2013). Other nutritional factors may affect cobalt absorption due to the formation of complexes
with certain organic anions (e.g., amino acids) present in foods.
Studies describing absorption of cobalt from the respiratory tract in humans are limited. Cobalt
levels in blood and urine of workers generally increase in proportion to inhalation exposure
levels to airborne cobalt dust and fumes, especially when workers were exposed to soluble
cobalt-containing particles (NTP 2014b, IARC 2006). The pattern of urinary excretion of cobalt
in workers exposed to less soluble cobalt oxide particles indicated a lower absorption rate and
longer retention time in the lungs. Deposition in the respiratory tract primarily depends on
20
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
particle size and breathing pattern (WHO 2006, ATSDR 2004). In general, particles larger than 2
µm tend to deposit in the upper respiratory tract due to higher airstream velocities and inertial
impaction. These particles are readily cleared through mucociliary action and swallowed.
Smaller particles escape inertial impaction and deposit in the bronchiolar or alveolar regions via
sedimentation and diffusion. Particles deposited in the respiratory tract may dissolve and be
absorbed into the blood or undergo phagocytosis or endocytosis by macrophages. In addition,
some nanoparticles can translocate rapidly from the lungs to the mediastinal lymph nodes and
bloodstream (Luyts et al. 2013). Recent in vitro studies with human lung cells show that
insoluble cobalt oxide particles (CoO or Co3O4) are readily taken up through endocytosis and are
partially solubilized at the low pH within lysosomes while soluble cobalt salts utilize cellular
transporters such as calcium channels or the divalent metal ion transporter to enter cells (Ortega
et al. 2014, Sabbioni et al. 2014, Smith et al. 2014, Papis et al. 2009). Controlled aerosol studies
using human volunteers show that about half of the initial lung burden of inhaled cobalt oxide
(Co3O4) particles may remain in the respiratory tract after six months (Bailey et al. 1989, Foster
et al. 1989).
Dermal absorption of cobalt was demonstrated in two studies that measured increased cobalt
concentrations in the urine of volunteers who immersed their hands in hard metal dust containing
5% to 15% cobalt for 90 minutes (Scansetti et al. 1994) or in a used coolant solution containing
1,600 mg/L cobalt for one hour (Linnainmaa and Kiilunen 1997). Cobalt also accumulated in the
fingernails of three cobalt-sensitive patients after immersing a finger in a cobalt salt solution for
10 minutes/day for 2 weeks (Nielsen et al. 2000). In vitro percutaneous absorption studies were
conducted with cobalt powder dispersed in synthetic sweat and applied to human skin mounted
on Franz diffusion cells (Larese Filon et al. 2009, Larese Filon et al. 2007, Larese Filon et al.
2004). The mean permeation flux was 0.0123 µg/cm2/hr, the lag time was 1.55 hr, and the
permeation coefficient was 0.00037 cm/hr. Median cobalt concentrations in the receiving phase
indicated that significantly more (~400 fold) cobalt penetrated damaged skin compared with
intact skin (Larese Filon et al. 2009). Cobalt was detected in its ionic form in both the donor and
the receiving phase. Significant amounts of cobalt also remained within the skin. These
experiments showed that skin absorption was closely related to the capacity of synthetic sweat to
oxidize metallic cobalt powder to soluble cobalt ions. No significant dermal absorption occurred
when cobalt was dispersed in a saline solution (Larese Filon et al. 2004).
Distribution and excretion
Cobalt occurs in most tissues of non-occupationally exposed people because it is a component of
vitamin B12. In humans, inorganic cobalt is distributed to liver, kidney, heart, and spleen with
lower concentrations found in bone, hair, lymph, brain, and pancreas (Paustenbach et al. 2013,
WHO 2006). Cobalt levels measured in blood and urine from various exposed populations
compared to control populations are discussed in Section 2-3 and summarized in Figure 2-1 and
Table B-1 In addition, several case-referent studies compared cobalt tissue levels in patients
dying from cancer with patients dying from other causes (see Appendix B). Cobalt chloride
administered intravenously (i.v.) or orally to human volunteers was distributed primarily to the
liver (Jansen et al. 1996, Smith et al. 1972). Whole body radioisotope scans (measured at various
times up to 1,000 days) following i.v. administration of inorganic cobalt indicated that 10% to
30% (mean 20%) was found in the liver (Smith et al. 1972). Cobalt levels in plasma declined
rapidly in this study due to rapid distribution to tissues and renal excretion; however, about 9%
This draft document should not be construed to represent final NTP determination or policy
21
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
to 16% of the administered dose was retained with a half-life of about 800 days. Measurements
of cobalt retention for up to 1,018 days indicated that about one fifth of the total body content
was present in the liver. Cobalt can also transfer to human milk and across the placenta (Rudge et
al. 2009, Wappelhorst et al. 2002). Most of the cobalt in plasma is bound to leukocytes or
plasma proteins with a maximum free fraction of 12%. Free cobalt is also taken up by red blood
cells via a membrane transport pathway shared with calcium (Simonsen et al. 2012, Simonsen et
al. 2011). Uptake of cobalt by red blood cells is practically irreversible because the ions bind to
hemoglobin and are not extruded by the calcium pump. Thus, it has been speculated that cobalt
partitions primarily into tissues with high calcium turnover and accumulates in tissues with slow
turn over of cells. Although elevated concentrations of cobalt have been reported in the liver and
kidney (oral or parental exposure) or lung (inhalation of insoluble particles), cobalt levels in the
body do not appear to increase in any specific organ with age (Lison 2015, Paustenbach et al.
2013, IARC 2006).
Renal excretion of absorbed cobalt is rapid over the first days but is followed by a second, slower
phase that lasts several weeks (Simonsen et al. 2012, IARC 2006). However, a small proportion
(~10%) is retained in the tissues with a biological half-life of 2 to 15 years. Controlled
experimental studies in humans indicate that 3% to 99% of an orally administered dose of cobalt
is excreted in the feces and primarily represents unabsorbed cobalt (WHO 2006). Fecal
elimination decreases as cobalt solubility increases. Following i.v. administration of cobalt
chloride to 6 volunteers, fecal elimination accounted for about 2% to 12% of the administered
dose while about 28% to 56% was eliminated in the urine after 8 days (Smith et al. 1972).
Valberg et al. (1969) reported similar results in subjects administered intramuscular injections of
cobalt and followed for 10 days (~6% excreted in feces and 58% in urine). Solubility and particle
size affect elimination following inhalation exposure (WHO 2006). Clearance of cobalt particles
from the lungs has been reported to follow three-phase kinetics (see Section 3.2.1). Large
particles are rapidly cleared from the upper airways via the mucociliary pathway, swallowed, and
eliminated in the feces. Urinary excretion of inhaled cobalt particles increases with time. Foster
et al. (1989) reported that following inhalation of cobalt oxide (Co3O4) particles, about 17% was
cleared mechanically to the gastrointestinal tract and eliminated in the feces within the first
week. After 6 months, about 33% of the initial lung burden was eliminated in the urine and about
28% was eliminated in the feces.
3.1.1
Experimental animals
The disposition of cobalt has been investigated in mice, rats, hamsters, guinea pigs, rabbits, dogs,
miniature swine, and baboons and show some similarities with human studies. These data are
briefly reviewed below. As in humans, cobalt as part of vitamin B12 is an essential micronutrient
in experimental animals. However, cobalt deficiency has been described in ruminants (e.g.,
sheep, goats, and cattle) raised in areas with very low cobalt (Yamada 2013). Cobalt supplements
were beneficial in these cases because cobalamin can be synthesized by gut bacteria and
absorbed.
Absorption
Cobalt absorption in experimental animals is highly variable and depends on the chemical form
of the compound, age of the animal, species, and nutritional status (NTP 2014b, WHO 2006,
Ayala-Fierro et al. 1999). In rats, cobalt chloride was absorbed more efficiently from the
22
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
gastrointestinal tract than insoluble cobalt oxide (Co3O4) (13% to 34% compared to 1% to 3%)
(NTP 2014b). Gastrointestinal absorption of soluble cobalt compounds was much lower in cows
(1% to 2%) and guinea pigs (4% to 5%) compared with rats. Cobalt absorption was 3% to 15%
greater in young rats and guinea pigs than in adults (Naylor and Harrison 1995). As observed in
humans, cobalt absorption was increased in iron-deficient rats (Thomson et al. 1971).
Inhalation studies of cobalt metal, cobalt oxides, or soluble cobalt salts in experimental animals
show that dissolved cobalt is absorbed rapidly from the lungs while a small percentage is
absorbed over several months (NTP 2014b, Leggett 2008, IARC 2006, NTP 1998, Kyono et al.
1992, IARC 1991). Cobalt particles are mechanically cleared by mucociliary action and
swallowed or phagocytized by macrophages. The fraction of the remaining lung content of cobalt
oxide (Co3O4) translocated to blood per day (i.e., dissolution of particles and absorption into the
blood) varied according to particle size, particle surface area, species, and time (Kreyling et al.
1991a, Andre et al. 1989, Bailey et al. 1989, Collier et al. 1989, Patrick et al. 1989). Initially,
translocation of the smaller particles (0.8 µm) ranged from about 0.4%/day in baboons to about
1.4%/day in the HMT (inbred strain) of rats. Initial translocation rates for the larger particles (1.7
µm) were lower in all species and ranged from about 0.2%/day in baboons to 0.6%/day in HMT
rats (Bailey et al. 1989). Translocation rates for higher density Co3O4 particles were about a
factor of 3 slower than for less dense particles (Kreyling et al. 1991a, Bailey et al. 1989).
Translocation rates reported by Bailey et al. (1989) showed a variety of different forms with
time, particularly for the smaller particles; this is discussed further in the following section
(Figure 3-2). Translocation of cobalt from the lung to the blood also was significantly faster in
younger rats compared with older rats (Collier et al. 1991).
This draft document should not be construed to represent final NTP determination or policy
23
06/05/15
24
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Figure 3-2. Rate of translocation of cobalt from lung to blood following inhalation of cobalt oxide particles
Source: (Bailey et al. 1989). Used with permission.
Dermal absorption of cobalt (applied as cobalt chloride) has been investigated in mice, guinea
pigs, and hamsters (Lacy et al. 1996, Kusama et al. 1986, Inaba and Suzuki-Yasumoto 1979).
Dermal absorption of cobalt applied to intact or acid-burned skin of mice was about 0.1% after
one hour but increased to 25% to 50% when applied to skin damaged by incision, abrasion, or
punctures (Kusama et al. 1986). In a similar study in guinea pigs, absorption of cobalt through
intact skin was less than 1% while absorption through abraded skin was about 80% 3 hours after
exposure (Inaba and Suzuki-Yasumoto 1979). Lacy et al. (1996) did not report the amount of
cobalt absorbed through the intact skin of hamsters but reported that small amounts of cobalt
were detected in urine 24 to 48 hours after application and that much of the metal was retained in
the skin after 48 hours. These authors also reported that uptake of cobalt by keratinocytes
exposed in vitro was about 5% of the dose.
Distribution and excretion
Absorbed cobalt is distributed rapidly to all tissues in experimental animals and is similar to that
in humans (NTP 2014b, WHO 2006). Edel et al. (1994) reported that tissue distribution
depended on dose, route of administration (oral versus parenteral), and time. Following oral
administration of cobalt compounds, the highest tissue concentrations generally occur in the liver
and kidney with lower amounts in the heart, spleen, muscle, bone, brain, pancreas, lung, and
This draft document should not be construed to represent final NTP determination or policy
25
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
gonads (Ayala-Fierro et al. 1999, Clyne et al. 1988, Gregus and Klaassen 1986, Bourg et al.
1985, Thomas et al. 1976, Hollins and McCullough 1971). Following single-dose parenteral
administration, some studies reported that concentrations were initially highest in the liver and
kidney but declined rapidly (Thomas et al. 1976, Hollins and McCullough 1971). However, Edel
et al. (1994) reported higher concentrations in the lung, large intestine, kidney, liver, and spleen
24 hours after a single i.v. injection of cobalt chloride. One hundred days after a single i.p.
injection, tissue distribution was affected by dose with higher concentrations in the spleen,
pancreas, and bone following the lower dose but mainly in bone following higher doses with
some accumulation in the heart.
Distribution of cobalt following inhalation exposure is similar to that observed for other routes
with the exception of greater retention in the lung for both soluble and insoluble cobalt (NTP
2014b, Patrick et al. 1994, Kyono et al. 1992, Collier et al. 1991, Bucher et al. 1990, Bailey et
al. 1989, Patrick et al. 1989, Kreyling et al. 1986, Kerfoot et al. 1975, Wehner and Craig 1972).
Long-term retention of insoluble cobalt particles and soluble cobalt salts deposited in the lung
shows wide interspecies variation and represents a potential continuing source of cobalt ion
release (Patrick et al. 1994, Kreyling et al. 1991a, Bailey et al. 1989). In addition, some particles
can translocate to the pulmonary interstitium where they are cleared from the lungs through the
lymphatic system (Pauluhn 2009). Nanoparticles also may penetrate the alveolar membrane and
distribute to extrapulmonary tissues via the circulation (Mo et al. 2008). The average size of the
long-term retention component in humans is greater than in experimental animals (Leggett 2008,
Bailey et al. 1989). Retention of insoluble cobalt oxide (Co3O4) particles (0.8 µm and 1.7 µm)
after 90 and 180 days are shown in Table 3-1. These data show that lung retention is generally
greater for larger particles than smaller particles and suggests temporal interspecies differences
in the rate of particle dissolution and absorption. However, the percentage of total body cobalt
content found in the lungs 30 and 180 days after exposure generally exceeded 90% in all species
for both particle sizes. In spite of considerable clearance from the lung, very little accumulated in
other tissues.
Table 3-1. Interspecies comparison of lung retention of cobalt oxide (Co3O4)
Lung retention (%)a
90 days
180 days
0.8 µm
1.7 µm
0.8 µm
1.7 µm
Human
64
75
45
56
Baboon
55
55
26
37
Dog, beagle
27
45
5.5
12
Guinea pig
49
46
8.3
15
Rat, HMT (1985)
5.2
20
1.3
8.0
Rat, HMT (1986)
5.3
18
1.2
7.2
Rat, F-344
14
25
4.7
9.2
Rat, Sprague-Dawley
8
39
1.0
15
Hamster, Syrian golden
21
35
3.4
12
Mouse, CBA/H
15
nd
2.8
nd
Species/strain
Source: (Bailey et al. 1989).
26
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
nd = no data.
a
Calculated as the fraction of lung content (measured as activity of 57Co) at 90 and 180 days relative to the lung activity at three
days after inhalation. The amount retained after three days was thought to be representative of the amount deposited in long-term
lung retention sites because, by this time, the rapid phase of mucociliary clearance should be complete.
Kreyling et al. (1991a) conducted a lung clearance study in baboons, dogs, and HMT rats using
Co3O4 particles (0.9 µm diameter) that were chemically similar to those used by Bailey et al. but
had a higher density (i.e., less porous) and a smaller specific surface area. In each species tested,
the denser 0.9 µm particles had higher lung retention after 90 and 180 days than the more porous
0.8 µm particles.
Bailey et al. (1989) and Kreyling et al. (1991a) also applied a simple dissolution model to predict
the diverse shapes of the time-dependent rate of cobalt translocation to blood from Co3O4
particles deposited in the lungs. This model was based on the assumption that the dissolution rate
is proportional to the specific surface area of the particle (surface area per unit mass). Since the
specific surface area increases as the particles dissolve, a high initial dissolution rate results in a
rapid increase in specific surface area and, in turn, causes an increase in the dissolution rate with
time. Thus, translocation will peak when another slow clearance mechanism is superimposed on
particle dissolution. A small fraction of the dissolved cobalt will not immediately translocate to
the blood but will be retained in the lungs and slowly released. The translocation rate was
defined in terms of two parameters: (1) the initial fractional absorption rate and (2) the fraction
of dissolved cobalt that is retained long-term in tissues (predicted as 1% to 10%). Although there
were some discrepancies between the curves predicted by the model and the observed
translocation rates (see Figure 3-2), overall, the model accounted remarkably well for the
different forms of translocation rates by varying the fractional dissolution rate and the long-term
retention fraction and suggested marked species differences in these parameters. The ratedetermining step for translocation was intracellular particle dissolution.
In an attempt to better understand the basis for the interspecies differences in the rate of Co3O4
absorption, species differences in lung retention and translocation (absorption) of soluble cobalt
chloride also was investigated (Patrick et al. 1994). The mean fraction of cobalt retained in the
lungs in the various test species administered cobalt chloride or cobalt nitrate (dog only)
(expressed as percent of initial body content) ranged from about 0.13% (hamster) to 1.2% (dog,
estimated value) after 100 days while the fraction retained in the whole body ranged from 0.35%
(hamster) to 3.2% (dog). Lung retention by species declined in the following order: dog > HMT
rat > guinea pig > baboon > F344 rat > hamster. These long-term retention values were lower
than the predicted values of 1% to 10% used in the model (see previous paragraph). The mean
fraction of cobalt retained in the lungs after 100 days in the various test species (expressed as
percent of cobalt remaining in the body after 100 days) ranged from 11.8% (baboon) to 60%
(HMT rat) with no significant accumulation in other organs with the exception of the trachea.
However, relative concentrations in the trachea showed no significant interspecies differences.
During the first week, 90% or more of the administered dose was cleared from the lung and was
similar to the pattern observed for i.v.-injected Co(NO3)2 in the same species (Patrick et al. 1994,
Bailey et al. 1989). These data suggest that interspecies differences in the time-dependent
absorption rates (i.e., translocation of dissolved cobalt from the lung to the blood) for inhaled
Co3O4 particles were not explained by differences in the fraction of dissolved cobalt retained
long-term in lung tissue. Kreyling et al. (1991b) also found little interspecies variation in pH
This draft document should not be construed to represent final NTP determination or policy
27
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
within alveolar macrophages; therefore, interspecies differences in translocation rates were not
explained by differences in phagolysosomal pH. Alternative explanations for these interspecies
differences could include a second long-term phase of lung retention as particles or as particle
fragments (Patrick et al. 1994).
A recent inhalation study with rats and mice exposed to cobalt metal showed that cobalt
concentrations increased with increasing exposure in all tissues examined; however, normalized
tissue burdens did not increase with increasing exposure (NTP 2014b). Cobalt tissue
concentrations (µg Co/g tissue) in male and female rats showed the following order: lung > liver
> kidney > femur > heart > serum > blood (NTP 2014b). Tissue cobalt burdens (µg Co/tissue)
showed a similar order with the exceptions that liver accumulated more cobalt than the lung, and
the heart accumulated more cobalt than the femur. At three weeks post-exposure in female rats,
cobalt concentrations were markedly reduced in blood, serum, and lung (no data were available
for other tissues). Tissue distribution data in mice were similar to that observed in rats but
concentrations in the femur and heart were similar to concentrations in blood and serum. These
data indicated that tissues tended to accumulate cobalt at concentrations greater than levels found
in the blood and serum and that cobalt was distributed to extra-pulmonary tissues.
Cobalt excretion occurs rapidly with the majority of the administered dose eliminated within
hours to a few days after exposure ceases (Paustenbach et al. 2013, Gregus and Klaassen 1986).
Cobalt is excreted in the urine, feces, and bile with similar excretion patterns reported for all
species studied (NTP 2014b, WHO 2006, ATSDR 2004). Most of the i.v.-injected dose of cobalt
chloride (~73% to 75%) was eliminated in the urine while smaller amounts were excreted in the
bile (2% to 5%) and feces (10% to 15%) (Ayala-Fierro et al. 1999, Gregus and Klaassen 1986).
Soluble cobalt compounds are cleared from the lungs at a faster rate than less soluble
compounds. The rate of urinary excretion correlates with the rate of translocation of cobalt from
the lungs to the blood while fecal excretion rates correlate with the rate of mechanical clearance
of cobalt particles from the lung (WHO 2006, ATSDR 2004). Following oral exposure, cobalt is
primarily excreted in the feces but the rate decreases as cobalt particle solubility increases (WHO
2006). However, species and sex differences in cobalt excretion rates have been reported. Cobalt
urinary excretion rates (µg/16 hr) in male rats were about two-fold higher than in females
exposed to various concentrations of cobalt sulfate for 13 weeks (Bucher et al. 1990). In another
study, mean urinary excretion rates of cobalt (administered as CoCl2 solution to the lungs or
inhaled as an aerosol) ranged from 0.002% of the initial body content per day in HMT rats to
0.026% per day in dogs (Patrick et al. 1994). Mean daily fecal excretion rates ranged from
0.0009% (dog) to 0.004% (HMT rat).
3.2
Toxicokinetics
Various toxicokinetic parameters of inorganic cobalt have been measured, and several
pharmacokinetic models have been developed that describe cobalt disposition in the body (Unice
et al. 2014, Paustenbach et al. 2013, Unice et al. 2012, Leggett 2008, ATSDR 2004). This
section provides a brief review of toxicokinetic data in humans (Section 3.2.1) and laboratory
animals (Section 3.2.2).
28
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
3.2.1
06/05/15
Humans
The kinetics of inhaled cobalt is determined by mechanical (mucociliary) clearance and by
translocation to blood and the lymphatic system (Figure 3-1) (ATSDR 2004). Foster et al. (1989)
calculated average translocation and mechanical clearance rates of inhaled cobalt oxide (Co3O4)
particles in four human volunteers. The ratio of translocation to mechanical clearance was about
5:1 for particle sizes of 0.8 and 1.7 µm. Inhalation studies in workers and volunteers exposed to
cobalt have shown that the elimination of insoluble cobalt metal or cobalt oxides (CoO or Co3O4)
from the lungs is multiphasic with reported half-lives of 2 to 44 hours, 10 to 78 days, and years
(NTP 2014b, WHO 2006, Mosconi et al. 1994a, Apostoli et al. 1994, Beleznay and Osvay 1994,
Newton and Rundo 1971). The elimination pattern was independent of the degree of exposure.
About 17% of the initial lung burden was eliminated within the first week while about 40% was
retained at 6 months after exposure (WHO 2006, Foster et al. 1989). These elimination phases
likely involve mucociliary clearance of cobalt particles from the tracheobronchial region,
macrophage-mediated clearance of cobalt particles from the lungs, and long-term retention and
clearance from the lung. The slower clearance with time likely reflects cobalt that is bound to
cellular components in the lung (WHO 2006, ATSDR 2004, Foster et al. 1989, Kreyling et al.
1986). Studies in human volunteers administered cobalt chloride by i.v. injection also show a
multiphasic elimination pattern (Holstein et al. 2015, Jansen et al. 1996, Letourneau et al. 1972,
Smith et al. 1972). These studies show that 36% to 44% of the administered dose is cleared with
a biological half-life of 6 to 12 hours, 45% to 56% is cleared with a biological half-life of 2 days
to 60 days, and 9% to 11% is cleared with a biological half-life of 600 to 800 days (Paustenbach
et al. 2013). Jansen et al. (1996) reported an apparent volume of distribution at steady state of 48
L that likely reflected initial accumulation in the liver (~50% of the administered dose).
Leggett (2008) developed a biokinetic model for inorganic cobalt that depicts recycling of cobalt
between blood and four systemic tissues (liver, kidneys, skeleton, and other soft tissues) and
transfer from blood to excretion pathways. The model assumes first-order kinetics, and
parameter values are expressed as transfer coefficients (fractional transfers per day) that were
largely derived from controlled humans studies. Unice et al. (2014, 2012) further refined this
model by incorporating different gastrointestinal absorption rates, adding compartments to
account for albumin-bound cobalt in intravascular and extravascular fluid, and accounting for
additional parameters such as total blood volume, red blood cell age, and urinary excretion rates.
The model was a reasonably good predictor of cobalt blood and urine concentrations measured in
male and female volunteers who ingested a cobalt supplement for 16 days to 3 months
(Tvermoes et al. 2014, Unice et al. 2014, Tvermoes et al. 2013).
3.2.2
Experimental animals
Lung clearance kinetics of cobalt particles include both mechanical transport and translocation
(Kreyling et al. 1991a, Bailey et al. 1989). Lung clearance of inhaled cobalt metal particles in
rats and mice showed a well-defined two-phase elimination profile following 3-month or 2-year
studies (NTP 2014b). The majority (> 95% in rats and > 82% in mice) of the deposited cobalt
was cleared rapidly (half-life of 1 to 5 days) while the remainder was cleared more slowly (halflives of ~20 to > 400 days) depending on the concentration and study duration. Lung steady-state
burdens were reached after approximately 6 months and were similar in rats and mice. Lung
cobalt burdens were well below the levels that would cause lung overload. Other studies showed
that interspecies differences in clearance patterns associated with mechanical transport and
This draft document should not be construed to represent final NTP determination or policy
29
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
translocation were not correlated. Initial mechanical clearance rates were typically 10- to 20-fold
greater in rodents than in other species, decreased monotonically with time, and were similar for
different particle sizes. In contrast, interspecies differences in translocation rates varied by 3- to
10-fold, remained constant or increased and then decreased with time, and were affected by
particle size (see Figure 3-2). Thus, in HMT rats, both rates were initially high, while in baboons
and humans both rates were low. Mice, hamsters, and F344 rats had high rates of mechanical
clearance but low to moderate rates of translocation while dogs had slow mechanical transport
but rapid translocation.
Thomas et al. (1976) reported that the whole-body half-life of 60CoCl2 administered by i.v.
injection was longer in the mouse (495 days) than in the rat (309 days), monkey (183 days), or
dog (180 days), but all were lower than values reported in humans (see Section 3.2.1). Other
studies in rats and dogs showed multiphasic first-order elimination kinetics following oral,
inhalation, or i.v. exposure (Table 3-2). These data indicate that soluble cobalt compounds are
cleared faster than cobalt metal in rats and that the cobalt oxide particle clearance in dogs during
the intermediate phase was proportional to particle size. Elimination of cobalt from the blood in
the recent NTP (2014b) study also indicated rapid and slow clearance phases; however, it was
not possible to fit the blood data to a two-compartment model due to the lack of early sampling
times. However, cobalt elimination half-lives estimated from blood concentrations on the last
day of exposure (2-week studies) and 3 weeks post-exposure were 9.2 to 11.1 days in female rats
and 4.1 to 7.3 days in female mice.
Table 3-2. Elimination half-lives for cobalt administered to experimental animals
Elimination T½
Reference
Species: exposure route
Compound(s)
Phase 1
Phase 2
Phase 3
(Ayala-Fierro et
al. 1999)
Male F344 rats: i.v.
CoCl2
1.3 hr
4.3 hr
19 hr
(Ayala-Fierro et
al. 1999)
Male F344 rats: oral
CoCl2
0.9a hr
4.6 hr
22.9 hr
(Menzel et al.
1989)
Male SD rats:
inhalation
CoCl2
1.8 hr
3.7–8.7b hr
−
(Kyono et al.
1992)
Male SD rats:
inhalation
Co metal
52.8c hr
52.8d hr
156c hr
172.8d hr
−
(Kreyling et al.
1986)
Male beagles:
inhalation
(endotracheal tube)
Co3O4
Co3O4 + CoO
Co(NO3)2
0.5 d
1−4 d
0.8 d
6–80e d
20−86e d
27 d
300–380 d
340–440 d
400 d
− = No data.
a
Absorption half-life.
b
Calculated from elimination rate constants of 0.188 h-1 (single exposure) and 0.08 h-1 (repeat exposure).
c
Lung.
d
Blood.
e
Half-lives were proportional to particle size.
3.3
Synthesis
Cobalt is absorbed from the GI tract, lungs, and skin and rapidly distributed throughout the body.
Absorption from the gastrointestinal tract is highly variable and is affected by the chemical form,
30
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
dose, age, formation of complexes with organic ions, and nutritional status. Soluble compounds
are absorbed to a greater extent than poorly soluble forms. Current biokinetic models assume GI
absorption of 20% to 45% for aqueous forms and 10% to 25% for solid forms. Studies in
experimental animals indicate higher absorption in young rats and guinea pigs than in adults
while studies in human volunteers indicate higher GI absorption in women than in men and may
reflect iron status. Cobalt absorption from the GI tract is higher in iron deficient humans and
experimental animals and suggests that cobalt and iron share a common uptake mechanism.
Cobalt levels in blood and urine of workers generally increase in proportion to airborne
concentrations. Although absorbed cobalt is distributed systemically, it does not accumulate in
any specific organ with age. Translocation rates of cobalt from the lung to the blood show
considerable interspecies variation with time and particle size with humans and baboons
generally having lower rates than dogs or rodents, and the whole-body half-life of cobalt was
longer in humans than in mouse, rat, monkey, or dog.
Cobalt excretion occurs rapidly with the majority of the administered dose eliminated within
hours to a few days after exposure ceases. Cobalt is excreted in the urine, feces, and bile with
similar excretion patterns reported for all species studied. Elimination in the feces primarily
represents unabsorbed cobalt while absorbed cobalt is eliminated in the urine. Toxicokinetic
studies indicate multiphasic elimination following inhalation of cobalt particles or i.v. injection
of cobalt chloride and generally show shorter elimination half-lives in experimental animals
compared to humans. Elimination half-lives reported for poorly soluble cobalt metal or cobalt
oxide particles from human lung ranged from 2 to 44 hours, 10 to 78 days, and years. These
elimination phases likely represent an initial rapid elimination from the tracheobronchial region
via mucociliary clearance, macrophage-mediated clearance, and long-term retention and
clearance. A similar pattern was reported in human volunteers given an i.v. injection of cobalt
chloride with about 40% cleared with a half-life of 6 to 12 hours, 50% cleared with a half-life of
2 to 60 days, and 10% cleared with a half-life of 600 to 800 days.
This draft document should not be construed to represent final NTP determination or policy
31
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
This Page Intentionally Left Blank
32
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
4 Human Cancer Studies
Introduction
The objective of the cancer hazard evaluation of cobalt and certain cobalt compounds (hereafter
referred to as cobalt) is to reach a preliminary level of evidence conclusion (sufficient, limited, or
inadequate) for the carcinogenicity of cobalt from studies in humans by applying the RoC listing
criteria to the body of evidence. In general, most of the human studies do not provide
information on the type(s) of cobalt compounds to which the subjects were exposed.
The steps in the cancer hazard evaluation, including the location of the discussion of these steps
in the document, are listed below.
1. Selection of the relevant literature included in the cancer evaluation (Section 4.1 and
Cobalt Protocol [NTP 2014c]).
2. Description of the study methods and characteristics (Appendix C.1, Tables C-1a-i) and
evaluation of study quality and other elements related to the utility of the studies to
inform the cancer hazard evaluation: Section 4.2 (cohort studies of lung cancer), Section
4.3 (case-control studies of esophageal, and other aerodigestive cancers (i.e., oral cavity,
laryngeal, and pharyngeal cancers), and Appendix C.2, Tables C-2a to C-2c).
3. Cancer assessment: Lung (Section 4.2.3), esophagus (Section 4.3.3), and other cancers
(Section 4.4).
4. Preliminary recommendation for the level of evidence of carcinogenicity (sufficient,
limited, or inadequate) of cobalt from human studies (Section 4.5).
The cancer hazard evaluation of cobalt primarily focuses on cancers of the lung, the esophagus,
and other aerodigestive cancers (i.e., oral cavity, laryngeal, and pharyngeal cancers) since these
are the only tissue sites that multiple studies evaluated. (For rationale, see Protocol: Methods for
Preparing the Draft Report on Carcinogens Monograph on Cobalt [“Cobalt Protocol”; NTP
2014c] and Tables 4-1 and 4-4). Because the occupational studies primarily reported on lung
cancer and the case-control studies reported on esophageal cancers and other aerodigestive
cancers, this section is organized by study design (following the selection of literature): cohort
studies and lung cancer are discussed in Section 4.2, case-control studies and esophageal cancer
in Section 4.3, and aerodigestive and other cancers (reported in both case-control and cohort
studies) are discussed in Section 4.4.
4.1
Selection of the relevant literature
Details of the procedures (such as the databases and literature search terms and screening
methods) used to identify and select the primary studies and supporting literature for the human
cancer evaluation are detailed in Appendix A and the cobalt protocol.
Primary epidemiologic studies were considered for the cancer evaluation if the study was (1)
peer reviewed; (2) provided risk estimates (or sufficient information to calculate risk estimates)
for certain cobalt compounds and human cancer, and (3) provided exposure-specific analyses for
cobalt or certain cobalt compounds at an individual level, or based on the authors’ report, cobalt
This draft document should not be construed to represent final NTP determination or policy
33
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
exposure was probable or predominant in the population, job, or occupation under study. Studies
of hip implants and prosthetic devices made from cobalt alloys and of radioactive cobalt were
excluded, as the extent of exposure to cobalt is often unclear and because of potential
confounding from the other compounds in the alloy (which are often metals) or radioactivity. In
general, cohort or case-control studies of populations with jobs, workplaces, or environmental
exposures in which cobalt exposure may have occurred (e.g., studies of hard-metal workers)
were excluded if a specific risk estimate for cobalt exposure alone was not reported as noted
above.
Biomarker studies of cobalt and cancer were included if they were conducted within defined
populations and provided risk estimates of cobalt levels and cancer. A series of clinical studies
that compared cobalt levels in hair of patients with cancer and referent groups or in tissues
(collected from autopsies) were identified and are summarized in tabular format in Appendix B,
Tables B-2 (hair) and B-3 (tissues). These studies did not provide information to calculate an
effect estimate, and most did not have defined methods for selecting the subjects and are not
included in the cancer hazard evaluation. Most of these studies measured numerous trace and
heavy metals.
Environmental studies of cobalt and cancer were included if they were conducted within defined
populations and provided risk estimates of cobalt levels and cancer. A total of four studies was
identified, two of which investigated the relationship of cobalt in air to breast (Coyle et al. 2006)
and lung (Coyle et al. 2005) cancer. The other two investigated the relationship between soil
levels of cobalt and cancer (McKinley et al. 2013, Kibblewhite et al. 1984). None of the studies
moved forward into the cancer hazard evaluation because they did not provide a risk estimate (or
sufficient information to calculate one) or exposure-specific analyses at the individual level.
4.2
Cohort studies and nested case-control studies reporting on lung cancer
This section provides an overview of the cohort and nested case-control studies (Section 4.2.1),
an overview of the adequacy of the studies to inform the cancer hazard evaluation (Section 4.2.2)
and an assessment of the evidence from the studies on the association between cobalt exposure
and lung cancer risk (Section 4.2.3).
4.2.1
Overview of the methodologies and study characteristics
For each of the reviewed cohort studies, detailed data on study design, methods, and findings
were systematically extracted from relevant publications, as described in the study protocol, and
into Table 4-1, Tables C-1a-g in Appendix C, and Table 4-2 in Section 4.2.2.
The available epidemiologic studies that satisfy the criteria for consideration in the cancer
evaluation consist of a series of occupational cohort or nested case-control studies conducted in five
independent populations. These include a cohort of female Danish porcelain painters; a cohort of
French electrochemical workers; and French cohorts of hard-metal workers, stainless and alloyed
steel workers; and Norwegian nickel refinery workers.
Tüchsen et al. (1996) reported on cancer incidence at multiple tissue sites among 1,394 female
porcelain painters employed in underglazing departments of two porcelain plate factories in
Denmark where cobalt-aluminate spinel and/or cobalt silicate was used, compared with top glaze
decorators in a department in one of the factories without cobalt exposure.
34
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Studies on the French electrochemical workers producing cobalt included two publications. The
first publication was on an historical mortality cohort and nested case-control study of lung
cancer among 1,143 cobalt production workers in a French electrochemical plant (Mur et al.
1987). This study included workers who had been employed for at least one year between 1950
and 1980. At this plant, cobalt was produced by etching roasted ore, followed by neutralization,
filtration, and electrolysis of the cobalt chloride solution. The manufacturing process also
included production of cobalt salts and oxides. The second publication was a re-analysis of the
cohort (N = 1,148), incorporating revised case-ascertainment and an extended period of followup (Moulin et al. 1993). The electrochemical worker cohort analyses reported findings for
trachea/bronchus/lung cancer, buccal cavity/pharynx, and larynx cancers (Mur et al. 1987); and
bronchus/lung, buccal-cavity/pharynx, larynx, esophagus, and brain cancers (Moulin et al. 1993).
Although both studied the same population, the original cohort is discussed because it contains
additional information (e.g., a nested case-control analysis) not included in the update.
Two publications reported on an overlapping population of hard-metal workers. The first was a
historical mortality cohort and nested case-control study of lung cancer among 7,459 workers at
10 hard-metal producing factories in France (Moulin et al. 1998) where activities also included
powder metallurgy processes. The second was a sub-study of lung cancer among 2,860 workers
in the largest hard-metal producing factory in France (the factory was included in the Moulin et
al. [1998] study, with an additional year of follow-up included) which also produced magnets
and stainless steel with cobalt, and cobalt powders by calcination and reduction of cobalt
hydroxide (Wild et al. 2000). This study also provided complete job histories.
A historical cohort and nested case-control study of stainless and alloyed steel workers and lung
cancer conducted in one factory in France (N = 4,897), which produced and cast stainless and
alloyed steel from cobalt, was also identified. Lastly, an incident nested case-control study of
213 cases of lung cancer among Norwegian nickel refinery workers was conducted to evaluate
whether exposure to cobalt (and other metals) could explain the elevated risk of lung cancer in
nickel workers.
In the two studies of electrochemical workers (Moulin et al. 1993, Mur et al. 1987), exposure
was assessed based on company records, which grouped workers into general service,
maintenance, and sodium production or cobalt production areas. Analysis was conducted for
“ever employment” in the cobalt production workshop, or for exclusive employment in this area.
Similarly, in the porcelain factories, exposure was based on company records, which grouped
workers into those who worked in departments with and without cobalt exposure (Tüchsen et al.
1996). Exposure to cobalt in the hard-metal factories, and the stainless and alloyed steel factory
was classified using a semi-quantitative job-exposure-matrix (JEM) developed by experts; the
nickel refinery workers were classified using this JEM which incorporated quantitative personal
measurements from the breathing zone.
All of the cohort and nested case-control studies reported on lung cancer alone, or lung cancer
and aerodigestive cancers, with only one of these reporting specifically about aerodigestive
cancers (i.e., buccal cavity/pharynx, and larynx cancers) (Mur et al. 1987) in relationship to
cobalt exposure. Only one study reported on multiple sites in relation to cobalt (i.e., cervix,
ovary, breast, and skin) (Tüchsen et al. 1996); thus, lung cancer is the only site with an adequate
database to contribute to the cobalt and cancer assessment.
This draft document should not be construed to represent final NTP determination or policy
35
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
The description of study methods and characteristics of each study is included in Appendix C,
Tables C-1a-g.
Table 4-1. Cohort and nested case-control studies of exposure to cobalt
Reference
(Tüchsen et
al. 1996)
(Mur et al.
1987)
(Moulin et al.
1993)
(follow-up)
Population
Cancer incidence cohort study
(SIR); Danish cancer registry
Cobalt-aluminate spinel; cobalt
silicate
1943–1992
N = 1,394 female
workers
874 exposed
520 unexposed
ICD-7: Lung (162.0, 162.1) and
16 other tissue/organ sites
Company records
French
electrochemical
workers
Historical mortality cohort
study (SMR) and nested casecontrol analysis (OR)
Mur et al. 1987
1950–1980
N = 1,143 males
Mur et al. 1987
ICD-8: All causes; trachea,
bronchus, and lung (162);
buccal cavity/pharynx/larynx
(140-149, 161)
Number of cobalt
production workers
NR
(Wild et al.
2000) (substudy of
largest plant)
French hard-metal
workers
Moulin et al. 1998
1945–1991
N = 7,459 men and
women; 68 cases and
180 controls
36
Exposed: Ever employed in two plate
underglazing factories
Unexposed: workers employed in a
cobalt-free department in one factory
Moulin et al. 1993
ICD-8: All causes; bronchus,
lung (162); brain (191)
Nested case-control analysis
(OR) and historical mortality
cohort study (SMR)
Moulin et al. 1998
ICD-8: Lung (162)
Wild et al. 2000
ICD-8: Lung (162)
Wild et al. 2000
1950–1992
N = 2,860 men and
women
Number of cobaltexposed only workers
NR
(Moulin et al.
2000a)
Exposure: Cobalt compounds,
assessment, metrics
Danish porcelain
painters
Moulin et al. 1993
1950–1988
N = 1,148
(Moulin et al.
1998) (multiplant)
Design and outcome (cancer
sites)
Production of cobalt, cobalt salts and
oxides. Company records classified
workers exclusively employed in one
of four work groups including cobalt
production workshop
Mur et al. 1987
Cohort analysis: Only/never
Nested case-control analysis:
Ever/never employed in cobalt
production
Moulin et al. 1993
Mortality SMR analysis
Only/never employed
Production of magnets, stainless steel,
and cobalt powders
“Other” cobalt exposure may have
included metallic and ionized cobalt
Semi-quantitative JEM
Moulin et al. 1998
Duration, intensity and cumulative
exposure
Wild et al. 2000
ever exposed
French stainless and
alloyed steel worker
cohort
Nested case-control analysis
(OR) within a historical
mortality cohort study
1968–1991
N = 4,897; 54 cases
and 160 controls
ICD-8: Lung (162)
Steel production and casting of
stainless steel, nickel, ferro-chromium,
and other ferroalloys in which iron,
chromium, nickel, and cobalt
compounds are used
Powder manufacture of metallic
powders
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference
Population
Design and outcome (cancer
sites)
06/05/15
Exposure: Cobalt compounds,
assessment, metrics
Semi-quantitative JEM
Duration, intensity, and cumulative
exposure
(Grimsrud et
al. 2005)
(methods
described in
Grimsrud et
al. 2003,
Grimsrud et
al. 2002)
4.2.2
Nickel refinery
worker cohort
1952–1995
N = 5,389; 213 cases
and 524 controls
Nested case-control analysis
(OR) within an incidence
cohort study; Norwegian
Cancer Registry
ICD NR: Lung
Cobalt present in raw materials and
intermediates in refinery and produced
electrolytically in an electrowinning
process
Breathing zone personal samples for
cobalt and nickel JEM
Quantitative cumulative exposure
Study quality and utility evaluation
This section provides an overview of the adequacy of the cohort and nested case-control studies
to inform the cancer hazard evaluation (see Appendix C for details on the assessment). This
assessment considers factors related to study quality (potential for selection and attrition bias,
information bias regarding exposure and outcome, and concern for inadequate analytical
methods, selected reporting, and inadequate methods or information to evaluate confounding)
and study sensitivity (e.g., such as adequate numbers of individuals exposed to substantial levels
of cobalt). The ratings for each of these factors are provided in Table 4-2 and a detailed
description of the rationale for the rating is provided in Appendix C.
No critical concerns for the potential for any of the biases (domains) were identified in the
available studies; thus, each may have some utility for evaluating potential cancer hazards. All of
the cohorts reported are relatively small or moderate sized and are, consequently, underpowered
due to few exposed cases or deaths. With one exception (Grimsrud et al. 2005), the cohort or
nested case-control studies included only very few cases exposed to cobalt alone, limiting their
statistical power to evaluate a modest risk of lung cancer (if it exists) from cobalt. In addition,
the level of exposure to cobalt alone in the cohort and nested studies was not defined with
enough detail (excepting Grimsrud et al. 2005) to explore exposure-response relationships. Table
4-2 depicts the overall assessment of the ability to inform the cancer evaluation based on the
overall utility of the studies, including potential for biases and study sensitivity.
The study of nickel refinery workers (Grimsrud et al. 2005) was considered to have the highest
quality because it had adequate numbers of exposed cases, evaluated cancer incidence,
incorporated quantitative assessments of exposure to cobalt, and had sufficient information on
potential confounders and co-exposures to incorporate these factors into analyses. However,
exposure to cobalt was highly correlated with nickel, which compromises the ability of the
statistical models to disentangle effects from the two exposures.
The remaining studies were also considered to have low/moderate ability to inform the cancer
hazard evaluation primarily because of more limited (semi-quantitative or qualitative) exposure
assessments, potential bias, and/or lower sensitivity. The major concern in the studies of hardmetal workers (Moulin et al. 1998, Wild et al. 2000) and stainless steel workers was potential
This draft document should not be construed to represent final NTP determination or policy
37
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
confounding from potential co-exposure to other lung carcinogens; this was also the case, but to
a lesser extent, for the electrochemical workers cohort. In the porcelain worker study (Tüchsen et
al. 1996), subcohorts of workers employed prior to 1981 when biomonitoring began and
exposure levels began to fall, would have contributed information about high exposures;
however, only estimates for the entire cohort were reported, potentially diluting the effect. No
relationship with duration of employment was found, but this was not reported by calendar
period. In the electrochemical workers cohort, concerns arose about the changing source of
outcome information from the first analysis (Mur et al. 1987) to the updated analysis (Moulin et
al. 1993). The change from use of medical records to death certificates, in combination with a
restriction to account for loss to follow-up in the foreign-born workers, reduced the estimate of
the risk in the follow-up study. In general, potential bias from these studies was in the direction
of the null, and they had limited sensitivity to detect an effect due to their small size or
inadequate information regarding level of exposure.
Table 4-2. Bias and quality summary for cohort and nested case-control studies
Qualitya
Exposure
Outcome
Confounding
methods
Adequacy of
analysis
Selective
reporting
Sensitivity
Integration
Utilityb
Selection
Bias
++
++
+++
++
++
+++
+
++
++
++
++
+
+++
+++
+
+
Hard-metal workers
Moulin et al. (1998)
++
++/+++
+++
+
+++
+++
++
++
Wild et al. (2000)
++
++/+++
+++
+
++
+++
++
++
+++
++
+++
+
+++
+++
++
++
+++
+++
+++
++
+++
+++
+++
+++
Citation
Porcelain painters
Tüchsen et al. (1996)
Electrochemical
workers
Moulin et al. (1993) (with
Mur et al. 1987)
Stainless and alloyed
steel workers
Moulin et al. (2000a)
Nickel refinery workers
Grimsrud et al. (2005)
NI = no information.
a
Levels of Concern for Bias and for Study Quality Rating – Equal column width for types of bias does not imply they have equal
weight (see appendix for description of terms): +++: low/minimal concern or high quality; ++: some concern or medium quality;
+: major concern or low quality; 0: critical concern.
b
Utility of the study to inform the hazard evaluation (see appendix for description of terms): ++++: high utility; +++: moderate
utility; ++: moderate/low utility; +: low utility; 0: inadequate utility.
4.2.3
Cancer assessment: Lung
The goal of the cancer assessment is to evaluate the evidence for the carcinogenicity of cobalt for
lung cancer. The conclusions regarding the assessment of study utility are brought forward, and
38
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
these are considered together with the evidence from the individual studies. Next, the evidence is
integrated across studies to reach a preliminary level of evidence conclusion to determine
whether there is credible evidence of an association between cobalt and lung cancer, and whether
such an observed association could be explained by chance, bias, or confounding.
Several of the guidelines developed by Austin Bradford Hill (Hill 1965) are relevant to the
evaluation of the level of evidence for this assessment, including the magnitude (strength) and
consistency of any observed associations across studies, evidence of an exposure-response
gradient, and temporality of exposure. The preliminary listing recommendation is provided in
Section 4.5.
Background information
Lung cancer is the third most common cancer in the United States, making up 13.5% of all new
cancers. The age-adjusted annual lung cancer rates (including trachea and bronchus) (per
100,000 males or females) in the United States from 2007 to 2011 (SEER 2015a) were
approximately 72.2 (male) and 51.1 (female) for incidence; and 61.6 (male) and 38.5 (female)
for mortality, with a 5-year survival rate of 16.8%. These data suggest that mortality and
incidence data are approximately comparable for informing the cancer assessment. Rates for new
lung and bronchus cancer cases have decreased on average 1.5% each year over the last 10 years;
and death rates have decreased on average 1.8% each year from 2002 to 2011. Incidence trends
and rates in European countries where all of the cohort studies were conducted are broadly
similar (Ferlay et al. 2013). For example, in the European Union, lung cancer incidence per
100,000 males is 66.3, and mortality is 56.4.
Latencies for solid tumors such as lung cancer are generally estimated to exceed approximately
20 years, but may vary considerably. Incidence rates of lung cancer generally increase after 50
years of age, and this cancer is most frequently diagnosed among people aged 65 to 74; the
median age at diagnosis is 70. None of the studies of cobalt and lung cancer included in this
review have indicated the sub-type(s) of lung cancer included in their analyses.
The single most important non-occupational risk factor for the development of lung cancer is
smoking. Other risk factors of concern include exposure to arsenic, asbestos, cadmium, silica,
chromates, nickel compounds, and polycyclic aromatic hydrocarbons, all of which are found in
cobalt manufacturing processes.
Evidence from individual studies
Based on the study quality evaluation, all six cohort and/or nested case-control studies reporting
on lung cancer and cobalt exposure were considered to have some utility for inclusion in the
cancer assessment. The findings of the individual studies are discussed below and presented in
Table 4-3. The available cohort and nested case-control studies of cobalt and lung cancer include
a cohort of Danish female porcelain painters, a cohort of French electrochemical workers, French
multi-centric cohort of hard-metal factory workers, a related cohort of workers from the largest
This draft document should not be construed to represent final NTP determination or policy
39
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 4-3. Evidence from cohort and case-control studies on lung cancer and exposure to cobalt
Reference, study-design,
location, and year
(Tüchsen et al. 1996)
Cohort
Copenhagen, Denmark
Factory 1: 1943–1992;
Factory 2: 1962–1992
(Mur et al. 1987)
Cohort
France
1950–1980
40
Population description
& exposure
assessment method
Danish porcelain
painters.
1394 total; 874 cobalt
exposed workers, 520
unexposed workers.
Exposure
assessment method:
company records
Electrochemical
workers
N=1143; number of
cobalt production
workers NR ~ 25% of
current staff at time
of publication
Exposure
assessment method:
company records
Exposure
category or
level
Exposed
cases/deat
hs
Risk estimate
(95% CI)
Co-variates
controlled
Lung (162 and 162.1)
All exposed
8
SIR
2.35 (1.09–
4.45)
Factory 1
(CoSilicate
from 1972)
3
1.6 (0.41–4.37)
Factory 2
(CoAlSpin
thru 1988)
5
3.25 (1.19–7.2)
Referents
7
1.99 (0.87–
3.94)
Age
Lung (162)
Cobalt
Production
4
SMR
4.66 (1.46–
10.64)
Age, year of
death
Comments, strengths, and weaknesses
Employment in factories/departments with
or without cobalt
Confounding:
Calculation of expected number of cancer
cases took five year age groups and
calendar periods in consideration. No HWE
Strengths:
Population exposed primarily to cobalt
compounds alone; only female population
with data on cobalt. No Healthy Worker
Effect apparent in cohort.
Limitations:
Small number of exposed cases. Sporadic
info on smoking, and no control for
smoking.
60% worked greater than 10 years; 75%
hired before 1975
Confounding:
SMR all cause mortality = 0.77 (P < 0.01);
no methods to control HWE; all cause
mortality for cobalt production = SMR 1.29
(0.86–1.87).
Strengths:
Cobalt production workers exposed
primarily to cobalt compounds; estimates of
cancer risk among those exclusively
employed in cobalt production.
Limitations:
Small number of exposed cases; high loss to
follow-up (20%); smoking history only
known on 30% of cases; co-exposure to
nickel and arsenic; significant HWE in this
cross-sectional cohort; no adjustments for
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference, study-design,
location, and year
Population description
& exposure
assessment method
Exposure
category or
level
Exposed
cases/deat
hs
Risk estimate
(95% CI)
Co-variates
controlled
06/05/15
Comments, strengths, and weaknesses
HWE or HWSE.
(Mur et al. 1987)
Nested case-control
France
1950–1980
Electrochemical plant
workers
Cases: 9; Controls: 18
Exposure
assessment method:
company records
(Moulin et al. 1993)
Cohort
France
Extended followup of
the Mur 1987 study
through 1988
Electrochemical
workers
Cohort 1 – N = 1,148;
Cohort II – N = 870;
number of cobalt
workers NR
Exposure
assessment method:
company records
Lung (162)
Ever
worked in
Co
production
4
OR
[4.0 (0.66–
24.4)]
None
Lung (162)
Exclusive
Cobalt
production,
Cohort I
3
OR
0.85 (0.18–2.5)
Exclusive
Cobalt
Production,
Cohort II
3
1.16 (0.24–3.4)
Ever
worked in
Cobalt
production,
4
0.88 (0.24–
2.25)
Age
60% worked greater than 10 years; 75%
hired before 1975
Confounding:
Cases (deaths from lung cancer) were
matched to controls (deaths from cause
other than cancer) for year of birth, age at
death, and smoking habits
Strengths:
Nested study reduces concern with strong
HWE in this cohort.
Limitations:
Small numbers with limited information on
exposures (only ever/never employment in
cobalt production department); no
description of control selection given only
30% had smoking data. No information on
differences in cases and controls by date of
hire or last employment to assess potential
HWSE.
NR, but likely similar to Mur 1987
Confounding:
No reported control for period effects,
duration, or and time since first exposure
Strengths:
Electrochemical plant workers exposed to
cobalt compounds, with exposure estimates
separated for workers in cobalt production.
Limitations:
Small number of exposed cases in overall or
sub-cohort; concern about outcome
misclassification; no consideration of
smoking; potential co-exposures to nickel
and arsenic. All cause mortality for the full
This draft document should not be construed to represent final NTP determination or policy
41
06/05/15
Reference, study-design,
location, and year
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Population description
& exposure
assessment method
Exposure
category or
level
Exposed
cases/deat
hs
Risk estimate
(95% CI)
Co-variates
controlled
Cohort I
Ever
worked in
Cobalt
production,
Cohort II
(Moulin et al. 1998)
Nested case-control
France
1968–1991
42
Workers in all 10
hard-metal factories
in France
Cases: 61; Controls:
180
Exposure
assessment method:
JEM
4
cohort was SMR = 0.85 (upper 95% CI =
0.95) due to large loss among foreign-born;
no internal analysis, nor any adjustments for
HWE or HWSE in this cross-sectional
cohort.
1.18 (0.32–
3.03)
Lung (162)
Exposure
level 2 to 9
15
OR
2.21 (0.99–4.9)
Exposure
intensity
trend
15
2.05 (0.94–
4.45)
Exposure
duration
15
2.2 (0.99–4.87)
Unweighted
cumulative
exposure
15
1.83 (0.86–
3.91)
Frequency
weighted
cumulative
exposure
trend
15
2.03 (0.94–
4.39)
Comments, strengths, and weaknesses
Unclear
which
variables
were
controlled in
the
multivariate
analysis for
cobalt alone
NR
Confounding:
mentioned the full list of IARC carcinogens,
but did not indicate if these were controlled
in the cobalt alone analyses
Strengths:
Data available for cobalt without tungsten
carbide; provided estimates for intensity,
duration, and cumulative exposure (both
weighted and unweighted); JEM validated
for atmospheric concentrations of cobalt;
incident cohort reducing the potential for
left truncation; internal analysis reducing
the impact of the reported HWE; and lagged
analysis.
Limitations:
Potential confounding by co-exposures
classified only as "ever/never" in the JEM.
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference, study-design,
location, and year
(Wild et al. 2000)
Cohort
France
1968–1992
Population description
& exposure
assessment method
Hard metal workers Largest plant in
France
2216 men and 644
women
Exposure
assessment method:
JEM
Exposure
category or
level
Exposed
cases/deat
hs
Risk estimate
(95% CI)
Co-variates
controlled
Lung (162)
Cobalt
except in
hard metals
15
SMR
1.95 (1.09–
3.22)
Unclear if
these are
crude
estimates,
Age
06/05/15
Comments, strengths, and weaknesses
NR
Confounding:
conducted separate smoking analyses
Strengths:
Incident cohort; ability to control for coexposures and smoking; no HWE; lagged
analysis.
Limitations:
Unclear if co-exposures or smoking are
controlled in the estimates for Co; external
analysis only presented; no exposure
metrics except for ever/never shown.
(Moulin et al. 2000a)
Nested case-control
France
1968–1992
Stainless and alloyed
steel workers
Cases: 54 (17 Co
exposed); Controls:
162 (67 Co exposed)
Exposure
assessment method:
JEM
Lung (162)
Exposed,
Crude
17
OR
0.64 (0.33–
1.25)
Exposed,
known
smoking
status,
Crude
12
0.62 (0.26–
1.46)
Exposed,
known
smoking
status,
smoking
adjusted
12
0.43 (0.16–
1.14)
Smoking
ever/never,
age, PAHs,
silica, gender
NR
Confounding:
Co correlated in a reported matrix with
Chromium and/or Nickel, and Iron, but
neither of these were included in the
multivariate analysis
Strengths:
Semi-quantitative JEM; exposure metrics
including duration and cumulative dose,
frequency weighted and unweighted; HWE
mitigated by use of internal analyses;
controlled for smoking; reported
information to assist in evaluating that
healthy worker survival bias was unlikely.
Limitations:
Models did not control for substances
This draft document should not be construed to represent final NTP determination or policy
43
06/05/15
Reference, study-design,
location, and year
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Population description
& exposure
assessment method
Exposure
category or
level
Exposed,
known
smoking
status, PAH,
silica, and
smoking
adjusted
(Grimsrud et al. 2005)
Nested case-control
Norway
1910–1995
Nickel refinery
workers
Cases: 213; Controls:
525
Exposure
assessment method:
JEM
Exposed
cases/deat
hs
12
Risk estimate
(95% CI)
Co-variates
controlled
related to Cobalt in the correlation analysis.
Known carcinogens had non-significant
ORs < 1.0, indicating that the study had low
sensitivity to detect an effect.
0.44 (0.17–
1.16)
Lung
Rise in OR
per (mg/m3)
× years,
Crude
NR
OR
1.3 (0.9–1.8)
Low (0.31–
29.5 µg/m3
× years
49
1.5 (0.6–3.8)
Med (29.7–
142 µg/m3 ×
years
73
2.4 (1–5.6)
High (144–
3,100 µg/m3
× years
82
2.9 (1.2–6.8)
Smoking
Lung
Rise in OR
per (mg/m3)
× years,
Crude
44
NR
0.7 (0.3–1.4)
Comments, strengths, and weaknesses
In µg/m3: High (144–3,100); Medium
(29.7–142); Low (0.31–29.5)
Confounding:
No multivariate estimates for the categorical
variable were able to be estimated due to
collinearity with nickel.
Strengths:
Quantitative cobalt levels reported based on
measurements from the breathing zone;
incident cases; internal analyses; relatively
large number of cases compared to other Co
studies.
Limitations:
Confounded by exposure to nickel;
collinearity was such that no estimates
controlled for other co-exposures was
possible.
Smoking,
nickel,
sulfuric acid
mists,
asbestos,
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference, study-design,
location, and year
Population description
& exposure
assessment method
Exposure
category or
level
Exposed
cases/deat
hs
Risk estimate
(95% CI)
Co-variates
controlled
06/05/15
Comments, strengths, and weaknesses
arsenic
Lung
Co
electrolysis
workshop,
0.03–2.2 yrs
23
1.6 (0.8–3)
Co
electrolysis
workshop,
2.3–11.8 yrs
44
2.8 (1.5–5)
Co
electrolysis
workshop
62
5.1 (2.9–9.1)
Smoking,
employment
in other
workshops
NR = Exposure levels or duration not reported.
This draft document should not be construed to represent final NTP determination or policy
45
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
factory in the multi-centric French hard-metal factory cohort, a cohort of French stainless and
alloyed steel workers, and a cohort of Norwegian nickel-refinery workers.
Porcelain painters
Tüchsen et al. (1996) reported a significantly increased risk of lung cancer in all exposed female
workers compared with the Danish female population (SIR = 2.35, 95% CI = 1.01 to 4.62, based
on 8 exposed cases). Factory-specific SIRs for lung cancer were also reported, indicating that
Factory 1, where cobalt aluminate-spinel was replaced by cobalt silicate in 1972, had a nonsignificantly elevated SIR of 1.6 based on 3 exposed cases (no CI provided); and that Factory 2,
where workers continued to be exposed to cobalt aluminate-spinel until 1989, had a significantly
elevated SIR of 3.25 based on 5 exposed cases. In addition, the authors reported an elevated SIR
of lung cancer in the referent group (SIR = 1.99, 95% CI = 0.80 to 4.11, 7 cases), similar in
magnitude to that found in the exposed group.
This study had low sensitivity to detect an effect because of (1) small numbers of exposed cases
in this relatively small cohort and (2) potentially combining workers with high and low
exposures together, which could dilute any effect and bias the results towards the null. In
addition, no lagged analyses were reported. A concern about differential selection also exists in
this study. The authors suggested that removal of records of ill persons was known to take place
in Danish manufacturing. The possibility of differential selection out of the cohort could have
resulted in an underestimation of the true incidence of lung cancer in this study.
An elevated lung cancer SIR, similar in magnitude to that reported in the exposed group was also
observed in the referents; a comparison of the exposed departments with the reference
department gave a relative risk ratio of 1.2 (95% CI = 0.4 to 3.8). The referents were reported to
be top glaze decorators employed in a department without cobalt exposure. Data from a previous
publication in this factory (Raffn et al. 1988) indicated an overlap of cobalt levels in referents
and exposed individuals, suggesting that the referents in the Tüchsen et al. paper were not
completely “unexposed.” Limited information regarding smoking and its potential relationship
with cobalt exposure was provided from two surveys of subsamples of workers (Prescott et al.
1992, Raffn et al. 1988). Based on a calculation of the weighted average of exposed and
unexposed respondents from both studies taken over the total sample size of the two studies, and
disregarding the specific cobalt compound to which workers were exposed, the smoking rate is
calculated to be 52% for exposed and 38% for referent women. The rate of smoking among
exposed women is close to that of skilled Danish women taken in 1982 (47%) and 1987 (55%);
and the rate of smoking in the referent group is similar to, but lower than, the rate in the general
population of Danish women (43% and 42% in these two years). This suggests that there may be
a non-smoking cause for the increased rate of lung cancer in the referent population, which might
be due either to misclassification of cobalt exposure, or to another unmeasured confounder. It is
also possible that cobalt-exposed workers are also exposed to the same unmeasured confounder,
although data from the substudy indicates that levels of silica, nickel, and dust were very low
based on air monitoring done in 1981 (Raffn et al. 1988). The porcelain painters cohort provides
inconclusive evidence for a carcinogenic effect of cobalt and lung cancer because of the finding
of similarly elevated levels of lung cancer among the referents.
The Tüchsen et al. (1996) study stands out from others in that it consists entirely of women.
Christensen et al. (1993a) conducted a cross-over study of oral administration of soluble and
46
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
insoluble cobalt compounds and found that there are clear differences in biological levels by
gender, with significantly higher urinary cobalt (higher uptake) levels and urinary excretion of
cobalt in females compared with males.
Electrochemical workers
Two publications reported on a cohort of cobalt production workers in a French electrochemical
plant (Moulin et al. 1993, Mur et al. 1987). Both findings are reported because the updated
follow-up (Moulin et al. 1993) not only introduced different case ascertainment methods than the
earlier analyses of the cohort (Mur et al. 1987), but also restricted analyses to account for large
loss to follow-up among foreign-born workers. The first paper reported a statistically
significantly increased SMR for lung cancer among the workers employed in cobalt production
only (SMR = 4.66, 95% CI = 1.46 to 10.64, based on 4 observed deaths) (Mur et al.). There was
large loss to follow-up and clear evidence of a healthy worker effect (HWE) (all-cause mortality
SMR = 0.77 [95% CI = 0.67 to 0.88]). However, in an internal matched analysis (matching
variables were year of birth, age at death, and smoking habits), no estimated odds ratio or
confidence interval was reported, but a crude calculation based on reported numbers of exposed
and unexposed cases and controls indicated an OR of 4.0 (95% CI = 0.7 to 24.4), indicating
internal consistency. However, in the follow-up of the cohort (Moulin et al. 1993) the SMR for
lung cancer among French-born workers exclusively employed in cobalt production was 1.16,
(95% CI = 0.24 to 3.40), based on 3 observed deaths. (The SMR for the entire cohort has lower
confidence because of high loss to follow-up and strong healthy worker effect among foreignborn workers). In addition, Moulin et al. reported a discrepancy in the number of observed cases
exclusively employed in cobalt production in the two analyses (e.g., Mur et al. [N = 4]; Moulin
et al. [N = 3]) due to differences in the methods used to ascertain cause of death. The Mur et al.
study used physicians’ medical records, whereas Moulin et al. (1993) used death certificates for
the years when they were available, and in the process, one exposed case was re-classified as
non-diseased; furthermore, during the extended follow-up, no additional lung cancer cases were
observed.
A further limitation of this study is its very weak consideration of risk factors for lung cancer,
particularly smoking status, and possible co-exposures in the cobalt production process to nickel
and arsenic. Mur et al. initially reported that smoking histories were available for 30% of
workers, and the authors matched cases and controls on smoking status; based on their table, a
crude OR of 4.0 was calculated. No explanation was given in that paper regarding the methods of
matching given the small percentage of workers with information on smoking status. However,
Moulin et al. did not address smoking in the analysis, but reported no excess of mortality from
circulatory and respiratory diseases, suggesting that smoking is unlikely to be higher in this
cohort than in the local French referent population.
The evidence from these electrochemical studies are inconclusive, based on the low sensitivity of
the Moulin et al. study to detect an effect, the lack of exposure metrics in both studies, and the
inability to control for confounding. The changed outcome classification across the two analyses
does not inspire confidence in the methods used in either study. The Mur et al. analysis was
consistent across the internal and external analyses, and was able to apply corrections to address
the HWE. The Moulin et al. analysis reduced power dramatically to detect an effect.
This draft document should not be construed to represent final NTP determination or policy
47
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
French hard-metal worker cohorts
The populations included in the two studies of cobalt exposure and lung cancer among hardmetal workers overlap, and both studies report either statistically significant elevated risks, or
borderline statistically significant risks, of lung cancer among those exposed to cobalt without
tungsten carbide. Moulin et al. (1998) first reported results from the multi-center study of 10
hard-metal factories in France. In the internal nested case-control analysis (Moulin et al. 1998),
based on 15 exposed cases, a borderline statistically significant increased risk of lung cancer was
associated with exposure (levels 2 to 9) to “cobalt alone or simultaneously with agents other than
tungsten carbide” compared with little or no exposure (levels 0 or 1) (OR = 2.21, 95% CI = 0.99
to 4.90). Regarding the presence of an exposure-response relationship, Moulin et al. reported
two-fold elevated trend tests (although not reaching statistical significance) based on 15 cases
across levels of exposure (OR = 2.05, 95% CI = 0.94 to 4.45), levels of duration (2.20, 95% CI =
0.99 to 4.87), cumulative weighted (1.83, 95% CI = 0.86 to 3.91), and cumulative un-weighted
doses (2.03, 95% CI = 0.94 to 4.39). Numbers of cases and category-specific OR estimates for
levels or categories of duration or cumulative dose were not provided. Wild et al. (2000) added
years of follow-up to the cohort from the largest factory included in the multi-center study and
found a statistically significant elevated SMR of lung cancer among those exposed to “cobalt
except in hard metals” based on the JEM (SMR = 1.95, 95% CI = 1.09 to 3.22). Wild et al.,
however, did not provide information on exposure-response relationships; and neither study
provided an examination of latency.
Moulin et al. (1998) and Wild et al. (2000) both measured and addressed co-exposures to 9
workplace lung carcinogens and smoking in analyses for cobalt-tungsten carbide. In both studies,
the JEM was used to assess exposure to other workplace carcinogens. Ever vs. never smoking
was obtained through interviews with cohort members, and their colleagues and relatives in the
Moulin et al. study and from occupational health department records in the Wild et al. study.
However, in both studies, it is unclear whether the analyses of cobalt alone included models for
adjusting for co-exposures to other carcinogens or smoking. In the Wild et al. (2000) study,
exposure to any IARC carcinogen without considering exposure to cobalt-tungsten carbide was
related to lung cancer (SMR = 2.05, 95% CI = 1.34 to 3.0).
Potential confounding from exposure to smoking is less of a concern in this study than potential
confounding from exposure to other carcinogens. There is no evidence from data presented to
indicate that exposure to cobalt alone and smoking was related. In addition, the low mortality
from smoking-related disease suggests a limited potential for confounding, as smoking is
unlikely to be more prevalent among the workers than in the overall population. In the French
cohort, mortality from chronic bronchitis and emphysema was low (SMR = 0.4, 95% CI = 0.05
to 1.44) and there was no consistent mortality pattern for other smoking-related cancers (e.g.,
larynx, bladder, buccal cavity/pharynx, and esophagus). In addition, as internal analyses are
usually assumed to be less affected by confounding from lifestyle factors (e.g., smoking) than
SMRs, the OR estimate from the multivariate model reported by Moulin et al. (1998) in the
internal analysis is likely to be the better estimate for cobalt and lung cancer from this cohort.
Due to the lack of information about control of carcinogenic co-exposures, confidence in the
finding is reduced.
48
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Stainless and alloyed steel cohort
No association between cobalt exposure and lung cancer was found in this study (Moulin et al.
2000a). In internal analyses of cobalt exposure based on the JEM in the stainless and alloyed
steel plant, Moulin et al. reported a crude OR of 0.64 (95% CI = 0.33 to 1.25), and an OR
adjusted for PAHs and silica OR of 0.58 (95% CI = 0.29 to 1.17) based on 17 exposed cases and
67 controls in 10-year lagged analyses. Similar findings were found among those with known
smoking habits (e.g., 12 cases and 36 controls). Moulin et al. (2000a) also reported significant
decreasing trends in duration, and frequency un-weighted and weighted cumulative dose for
workers with known smoking habits. (The overall cohort SMR for smoking and lung cancer was
5.37 [95% CI = 1.74 to 12.53] for those working less than 10 years). ORs adjusted for smoking
were all less than 1.0 (Moulin et al.). It is likely that non-differential exposure misclassification
was introduced into the exposure assessment because some job periods of cases or controls went
back many decades, yet exposure was assessed based on memories of processes and exposures of
current workers or reports in the literature, as historical exposure measurements were lacking.
Models were reported controlling for PAHs and silica, none of which made any material
difference; however, in the correlation matrix neither of these was related to cobalt exposure.
Exposure to nickel and/or chromium was related to cobalt exposure, although these were not
included in the cobalt model. However, these exposures were also not associated with lung
cancer risk in these data.
In this study, chromium and/or nickel and asbestos, all lung carcinogens classified by RoC and
IARC, were found to be unrelated to lung cancer, decreasing the confidence in this study and in
the findings for cobalt. Only exposure to PAHs and silica were statistically significantly related
to lung cancer along with increasing trends not confounded by smoking.
Misclassification of exposure in this study, its inability to control for the appropriate confounders
correlated with cobalt, and the negative findings for lung cancer and other known lung
carcinogens (e.g., nickel, chromium, asbestos) suggest little confidence in the evidence put forth
in this study.
Norwegian nickel refinery workers
The Grimsrud et al. (2005) cancer incidence study of nickel and lung cancer in a Norwegian
nickel refinery was conducted to determine if cobalt or other potential carcinogens could explain
the elevated risks of lung cancer in nickel workers. The authors reported that the cobalt variable
could not be retained in the full model in its categorical form due to collinearity (all individuals
exposed to nickel were also exposed to cobalt, although the correlation between cobalt and
nickel was reported as r = 0.63); however, the positive exposure-response effect noted for the
continuous cobalt variable adjusted only for smoking changed sign when smoking and coexposures (nickel, arsenic, asbestos, and sulfuric acid mists) were controlled. The smokingadjusted rise in OR per mg/m3 × year was 1.3 (95% CI = 0.9 to 1.8), which was reduced to 0.7
(95% CI = 0.3 to 1.4) after adjustment for occupational co-exposures. The categorical ORs
adjusted only for smoking were: low exposure (0.31 to 29.5 μg/m3) based on 49 cases, OR = 1.5
(95% CI = 0.6 to 3.8); medium exposure (29.7 to 142 μg/m3) based on 73 cases, OR = 2.4 (95%
CI = 1.0 to 5.6); and high exposure (144 to 3,100 μg/m3) based on 82 cases, OR = 2.9 (95% CI =
1.2 to 6.8). No value for trend was reported for the smoking-adjusted variable. However, the
This draft document should not be construed to represent final NTP determination or policy
49
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
fully adjusted model for this cobalt variable (including smoking as well as all co-exposures)
could not be calculated due to collinearity.
The authors reported that cobalt typically amounts to 4% to 15% of the total nickel, except in the
cobalt electrolysis process where cobalt levels are triple the amount of nickel levels. This process
is included in hydrometallurgical production, for which results are reported by duration of work.
Strong gradients were found by duration of work in the hydrometallurgical production
department with a 5-fold increase in the OR for 12 or more years (OR = 5.1, 95% CI = 2.9 to
9.1) based on 62 exposed cases, with the linear trend (per 10 years) (OR = 1.7, 95% CI = 1.4 to
2.1). However, no analyses were provided to help separate effects of exposure to cobalt and
nickel.
Although the design of this study was of high quality, due to the collinearity with exposure to
nickel, this study cannot separate out the effects of cobalt and nickel on lung cancer and thus the
findings from the study are unclear.
Integration of evidence across studies
While almost all the cohort studies reported approximately a doubling of the risk of lung cancer
mortality or incidence from exposure to various cobalt compounds, it is unclear that the excess
lung cancer was due to exposure specifically to cobalt, because 1) it was not possible to rule out
confounding by carcinogenic co-exposures, or 2) other complications prevented a clear
interpretation of a cobalt effect.
The Danish porcelain painters study showed similarly elevated risks of lung cancer in both the
exposed and unexposed workers, and could not control directly for smoking. Findings from the
French electrochemical workers cohort were based on two papers analyzing the same cohort
using different methods to ascertain cancer, and publishing conflicting results – the first
indicated a significantly elevated risk of lung cancer based on 4 exposed cases, and the second
showed virtually no differences in risk of lung cancer among the exposed and unexposed
workers based on 3 exposed cases in a subset of workers born in France. In two French studies
of hard-metal workers, measures of cobalt exposures were likely mixed with other carcinogens
and the methods did not clearly indicate whether these were controlled in the analyses. Although
an exposure-response relationship between cobalt exposure and lung cancer was observed in the
Norwegian nickel refinery workers study, risk estimates could not be calculated in models
controlling for other co-exposures because nickel and cobalt were highly correlated. However,
in this study a significant trend was reported with increasing duration of employment in
workshops where cobalt concentrations tripled those of nickel, controlled for employment in
other workshops and smoking. Confounding by smoking was considered in each of the studies to
varying degrees, and smoking either did not reduce the risk estimates materially when it was
controlled, or was unlikely to materially reduce the risk estimates in studies where there was only
auxiliary information.
50
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Figure 4-1. Forest plot showing risk ratios (SIR, SMR or OR as noted) and 95% CI for epidemiological studies
of cobalt exposure
This draft document should not be construed to represent final NTP determination or policy
51
06/05/15
4.3
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Case-control studies
This section provides an overview of the case-control studies (Section 4.3.1), an overview of the
adequacy of the studies to inform the cancer hazard evaluation (Section 4.3.2) and an assessment
of the evidence from the studies on the association between cobalt exposure and esophageal
cancer risk (Section 4.3.3).
4.3.1
Overview of the methodologies and study characteristics
The available epidemiologic studies that satisfy the criteria for inclusion in the review consist of
two population-based case-control studies of metals in biological tissues of cancer cases (lung,
esophageal, oral cavity, and laryngeal cancers) and controls published in the literature between
1986 and 2012 (Table 4-4). Both of these studies (O'Rorke et al. 2012, Rogers et al. 1993) were
initiated from an interest in the role of metals in the etiology of cancer, and specifically metals
derived from nutritional sources. Detailed data on study design, methods, and findings were
systematically extracted from relevant publications, as described in the study protocol, into Table
4-5, Tables C-1h,i in Appendix C, and Table 4-6 in Section 4.3.2.
Table 4-4. Case-control biomarker studies of exposure to cobalt
Reference
(Rogers et al.
1993)
(O'Rorke et al.
2012)
Design and population
Outcome
Exposure: Cobalt
compounds, assessment,
metrics
Population-based casecontrol biomarker
study
Western WA state
USA
1983–1987
501 cases (153
laryngeal, 73
esophageal, 359 oral
cavity cancers)/434
controls
ICD-O
Larynx (140.0–141.9)
Esophagus (143.0–146.9)
Oral cavity (148.0–150.9;
161.0–161.9)
Source and type of
compounds unknown
Population-based casecontrol biomarker
study
Ireland
FINBAR* study
2002–2004
137 cases/221 controls
ICD not reported
Esophagus
Barrett’s esophagus
(metastatic precursor to
esophageal cancer)
Source and type of
compounds unknown
Cobalt levels in toenails
measured
Tertiles (ppm)
Cobalt levels in toenails
measured
Tertiles (log transformed
- cut points μg/g)
*FINBAR = Factors Influencing the Barrett’s Adenocarcinoma Relationship.
4.3.2
Study quality and utility evaluation
This section provides an overview of the adequacy of the cohort and nested case-control studies
to inform the cancer hazard evaluation (see Appendix C for details on the assessment). This
assessment considers factors related to study quality (potential for selection and attrition bias,
52
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
information bias regarding exposure and outcome, and concern for inadequate analytical
methods, selected reporting, and inadequate methods or information to evaluate confounding)
and study sensitivity (e.g., such as adequate numbers of individuals exposed to substantial levels
of cobalt). The ratings for each of these factors are provided in Table 4-5 and a detailed
description of the rationale for the rating is provided in Appendix C.
Both of the case-control studies of cobalt in toenails have either low/minimal or some concern
for most biases except for exposure assessment and sensitivity. Their overall low utility to
inform the cancer hazard evaluation, however, is due to the potentially irrelevant window of
exposure and potential for reverse causation. However, although exposure was assessed after the
disease process began, in most cases it represents at least some pre-diagnosis exposure, but not
pre-cancer exposure as the latency period of both esophageal cancer and Barrett’s esophagus is
of long duration (Butt and Kandel 2014). Rogers et al. conducted stratified analyses on tumor
stage and time of diagnosis that may be of help to evaluate the potential for reverse causality.
Table 4-5. Bias and quality summary for case-control studies
Outcome
Confounding
methods
Adequacy of
analysis
Selective
reporting
Sensitivity
Integration
Utilityb
Exposure
Qualitya
Selection
Biasa
Rogers et al. (1993)
+++
+
+++
+++
+++
+++
+
+
O'Rorke et al. (1993)
++
+
+++
+++
+++
+++
+
+
Citation
a
Levels of Concern for Bias and for Study Quality Rating – Equal column width for types of bias does not imply they have equal
weight (see appendix for description of terms): +++: low/minimal concern or high quality; ++: some concern or medium
quality; +: major concern or low quality; 0: critical concern.
b
Utility of the study to inform the hazard evaluation (See appendix for description of terms): ++++ high utility; +++: moderate
utility; ++: moderate/low utility; +: low utility; 0: inadequate utility.
4.3.3
Cancer assessment: Esophageal cancer
Background information
Esophageal cancer is a relatively rare cancer, ranking as the eighteenth most common cancer in
the United States, making up 1.1% of all new cancers. The age-adjusted annual rates of
esophageal cancer (per 100,000 males or females) in the United States from 2007 to 2011 (SEER
2015b) were approximately 7.7 (male) and 1.8 (female) for incidence; and 7.5 (male) and 1.6
(female) for mortality, with a 5-year survival rate of 17.5%. Like lung cancer, these data suggest
that mortality and incidence data are approximately comparable for informing the cancer
assessment. Incidence trends and rates in European countries where all of the cohort studies were
conducted are broadly similar (Ferlay et al. 2013); and in the European Union the annual
incidence of esophageal cancer is 8.4 and the annual mortality rate is 7.0 (Cancer Research UK
2014). Evaluations of esophageal cancer risk factors have reported that sufficient evidence exists
for x-and gamma-radiation, alcoholic beverages, betel quid, tobacco smoking, and smokeless
tobacco; limited evidence exists for dry-cleaning, mate drinking, pickled vegetables, rubber
production industry, tetrachloroethylene exposures, red and processed meats, and high
This draft document should not be construed to represent final NTP determination or policy
53
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
temperature drinks. The sub-types of esophageal cancer, esophageal adenocarcinoma (EAC) and
esophageal squamous-cell cancer (ESCC), however, have distinct risk factors and trends. EAC,
with risk factors being white race, increasing age, body fatness, and male gender, is the
predominant histological type among men, while for women, ESCC is more common and rates
are still increasing in several European countries. Unlike esophageal squamous-cell carcinoma,
alcohol is not a risk factor for either Barrett’s esophagus or for esophageal adenocarcinoma
(Freedman et al. 2011, Anderson et al. 2009, Kubo et al. 2009); however, smoking is a risk
factor for both subtypes and Barrett’s esophagus (Cook et al. 2010).
Barrett’s esophagus is a condition of intestinal metaplasia in which tissue that is similar to the
lining of the intestine replaces the tissue lining of the esophagus. The prevalence of Barrett’s
esophagus is estimated to be between 1.6% and 6.8% (Gilbert et al. 2011), although a more
precise estimate is not possible as many patients are asymptomatic, and its natural history has
been difficult to assess. Barrett’s esophagus has an extended latency period prior to progressing
to cancer (Butt and Kandel 2014). A recent meta-analysis of studies reports incidence rates for
the development of esophageal cancer in nondysplastic Barrett’s esophagus of 0.33% per year
and 0.19% for short-segment Barrett’s esophagus (Desai et al. 2012). About 5% of patients with
esophageal adenocarcinoma have a pre-cancer diagnosis of Barrett’s esophagus (Corley et al.
2002); but its presence conveys a 30- to 40-fold increased risk of esophageal carcinoma (Sharma
2004). As incidence of esophageal adenocarcinoma has increased more than six-fold in the last
decade, investigations of the risk factors for Barrett’s esophagus have been of interest (Jemal et
al. 2013). Barrett’s esophagus incidence increases with age; the prevalence among non-Hispanic
whites is 6.1% compared to 1.7% among Hispanics and 1.6% among blacks; and the
male/female ratio is about 2:1 (Abrams et al. 2008), similar to esophageal cancer.
Evidence from individual studies
Both of the case-control studies (O'Rorke et al. 2012, Rogers et al. 1993) compared cobalt in
toenails of cases of esophageal cancer and population-based controls. O’Rorke et al. limited their
analysis to esophageal adenocarcinoma, while no histologic information was provided by Rogers
et al., thus it is likely that the Rogers et al. study included both subtypes in unknown proportions.
Findings are presented in Table 4-6.
54
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Table 4-6. Evidence from studies of aerodigestive cancers and exposure to cobalt
Reference, study-design,
location, and year
(Rogers et al. 1993)
Case-control
Western WA state,
USA
9/1/83–2/28/87
Population description &
exposure assessment
method
Population based study
of aerodigestive
cancers, USA
Cases: N = 507; N =
153 laryngeal, N = 73
esophageal, N = 359
oral cavity cancers;
Controls: N = 434
Exposure assessment
method: personal
monitoring
Exposure
category or
level
Exposed
cases/
deaths
Risk estimate
(95% CI)
Co-variates
controlled
Esophagus (143.0-146.9)
< 0.05
92
OR = 1.0
0.05−0.17
127
2.4 (0.8–7.2)
> 0.17
66
9 (2.7–30)
Age, sex, smoking
(pack-years),
alcohol, drink years,
beta-carotene,
mg/day, energy
intake, kcal/day,
ascorbic acid
mg/day
Larynx (140.0-141.9)
< 0.05
114
OR = 1.0
0.05-0.17
168
2 (1–3.8)
> 0.17
62
1 (0.4–2.6)
Age, sex, smoking
(pack-years), energy
intake, kcal/day,
beta-carotene,
mg/day, ascorbic
acid mg/day,
alcohol, drink years
Comments, strengths, and
weaknesses
Tertiles of cobalt in toenails;
highest level 0.17 ppm
Confounding:
Nutrients in the model did not
greatly confound the relationship
between exposure and disease, but
inclusion resulted in ORs closer to
the null. ORs for esophageal
cancer significantly elevated for
iron and calcium
Strengths:
Population-based study;
histologically confirmed cancers;
cases and controls from same
source population; matching on
key likely confounders
Limitations:
Toenail levels pre-diagnostic, but
reverse causation can't be ruled out.
No information provided about
correlation of cobalt with other
trace elements.
Oral Cavity (148.0-150.9; 161.0-161.9)
< 0.05
135
OR = 1.0
0.05−0.17
190
1.5 (0.9–2.6)
> 0.17
92
1.9 (1–3.6)
Age, sex, smoking
(pack-years),
alcohol (drink
years), energy
intake (kcal/day),
ascorbic acid
(mg/day), betacarotene (mg/day)
This draft document should not be construed to represent final NTP determination or policy
55
06/05/15
Reference, study-design,
location, and year
(O'Rorke et al. 2012)
Case-Control
All Ireland (Republic
and Northern)
3/2002–12/2004
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Population description &
exposure assessment
method
All Ireland population
based study of
esophageal cancer and
Barrett's esophagus
Cases: N = 137 for
esophageal cancer, N =
182 for Barrett’s
esophagus; Controls: N
= 221
Exposure assessment
method: personal
monitoring
Exposure
category or
level
Exposed
cases/
deaths
Risk estimate
(95% CI)
Co-variates
controlled
Esophageal cancer
< -5.4824
34
OR = 1.0
≥ -5.4824
39
1.06 (0.57–1.98)
≥ -4.4705
52
1.54 (0.84–2.85)
Age, sex, GI reflux,
education, H. pylori
infection, location,
smoking
Trend-test P-value: 0.16
Esophageal cancer
< -5.4824
34
OR = 1.0
≥ -5.4824
39
1.13 (0.64–1.99)
≥ -4.4705
52
1.54 (0.9–2.68)
Age, sex
Trend-test P-value: 0.11
Barrett's Esophagus
< -5.4824
55
OR = 1.0
≥ - 5.4824
54
1.08 (0.55–2.1)
≥ -4.4705
64
1.97 (1.01–3.85)
Age, sex, GI reflux,
H. pylori infection,
smoking habits,
energy intake,
location
Trend-test P-value: 0.05
Barrett's Esophagus
< 5.4824
55
OR = 1.0
≥ 5.4824
54
0.97 (0.59–1.59)
≥ 4.4705
64
1.18 (0.72–1.93)
Age, sex
Comments, strengths, and
weaknesses
Average (μg/g ) ± SD : cases =
0.02 ± 0.06; controls = 0.02 ± 0.04.
Range: cases = 0.002– 0.60;
controls = 0.002– 0.47
Confounding:
Unadjusted model almost identical
results to the age and sex adjusted
model, other metals measured
included selenium, chromium,
zinc, mercury, cerium. No
correlation with cobalt reported.
Not in models.
Strengths:
Population based; histologically
confirmed cancer; data on broad
range of co-exposures and covariates collected.
Limitations:
Differences in sources of cases and
controls in No Ireland and Rep of
Ireland may introduce some
selection bias; low participation
rate in controls, especially in Rep
of Ireland; no correlation
information of cobalt with other
trace elements provided.
Trend-test P-value: 0.5
(Mur et al. 1987)
Cohort
France
1950–1980
56
Electrochemical
workers
N=1143; number of
cobalt production
workers NR ~ 25% of
current staff at time of
Buccal cavity, pharynx, larynx (140-149, 161)
Cobalt
Production
2
SMR
3.36 (0.29–
10.29)
Age, year of death
60% worked greater than 10 years;
75% hired before 1975
Confounding:
SMR all cause mortality = 0.77 (P
< 0.01); no methods to control
HWE; all cause mortality for cobalt
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference, study-design,
location, and year
Population description &
exposure assessment
method
publication
Exposure assessment
method: company
records
Exposure
category or
level
Exposed
cases/
deaths
Risk estimate
(95% CI)
Co-variates
controlled
06/05/15
Comments, strengths, and
weaknesses
production = SMR 1.29 (0.86–
1.87).
Strengths:
Cobalt production workers exposed
primarily to cobalt compounds;
estimates of cancer risk among
those exclusively employed in
cobalt production.
Limitations:
Small number of exposed cases;
high loss to follow-up (20%);
smoking history only known on
30% of cases; co-exposure to
nickel and arsenic; significant
HWE in this cross-sectional cohort;
no adjustments for HWE or
HWSE.
This draft document should not be construed to represent final NTP determination or policy
57
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Western Washington state study of aerodigestive cancers
Rogers et al. (1993) reported elevated odds ratio for esophageal cancer for those with the highest
levels (≥ 0.17 ppm) of cobalt concentration in toenails compared to those with the lowest level (<
0.05 ppm) of cobalt (OR = 9.0, 95% CI = 2.7 to 30.0). The OR was elevated but not significant
for those with medium levels (0.05 to 0.17 ppm) of cobalt concentration compared to those with
low levels (OR = 2.4, 95% CI = 0.8 to 7.2). The exposure-response test for trend was significant
(P < 0.001). It is not possible to comment on the distribution of levels of cobalt in the cases
compared to the controls, as cases and controls are combined across exposure levels.
Confounding from known risks factors for esophageal cancer can reasonably be ruled out, except
for the metals, as none have previously been associated with esophageal cancer. In this study, the
risk of esophageal cancer was also associated with elevated levels of calcium and iron. Smoking
and alcohol use were controlled in the multivariate models along with age and gender, energy
intake, beta-carotene and ascorbic acid; however, while cases were less educated than controls,
this variable was not included in the model. Neither beta-carotene nor ascorbic acid confounded
the relationships between cobalt and esophageal cancer, but the authors included these two
nutrients in the logistic model as it reduced the ORs slightly, raising the concern that the model
estimates may have been over-controlled, biasing them slightly towards the null. Co-exposures
from other metals were not reported or considered in the analysis of cobalt, and no correlations
among the metals were reported.
The source of the cobalt exposure is unknown. When cobalt in nail tissue was expressed as a
continuous variable, there were no associations between nail concentration of cobalt and dietary
intake of foods high in cobalt (e.g., meat) suggesting that diet does not explain the elevated
levels of cobalt in cases. Although occupational histories using questionnaires were collected in
this study, no exposure assessment or analyses were done specifically for exposure to cobalt.
Although the Rogers et al. study provides some evidence of an association, the analysis of a
single sample of toenail clippings collected near the time of diagnosis, with no accompanying
data on potential sources of cobalt from the environment or occupational exposure, limits the
utility of the study. Reverse causality due to trace element deposition in nails influenced by
factors associated with cancer (e.g., weight loss, age, gender, changes in diet and smoking and
alcohol consumption) (Slotnick and Nriagu 2006, Hunter et al. 1990) may be a possibility.
However, the authors stated (data not shown) that elevated risks of esophageal cancer were
found in individuals with in situ/localized tumors as well as those with regional/distant tumors,
and no significant differences were found in stratified analysis by the time from diagnosis to
tumor (< 7 months or ≥ 7 month), suggesting that reverse causality may not be a concern.
Finbar study – Ireland
O’Rorke et al. (2012) reported a non-significant elevated risk of esophageal adenocarcinoma
among those with the highest cobalt levels (OR = 1.54, 95% CI = 0.84 to 2.85). In addition, they
reported a significantly increased risk of Barrett’s esophagus among participants with higher
toenail concentrations of cobalt (≥ -4.4705, log transformed values equivalent to ≥ 0.011 μg/g)
(OR = 1.97, 95% CI = 1.01 to 3.85), with a significant (P = 0.05) linear test for trend. Both of
the estimates were adjusted for age, sex, smoking, location (Northern Ireland or Republic),
energy intake, gastro-esophageal reflux, and H. pylori infection. O’Rorke et al. reported no
58
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
05/07/15
information regarding the correlation of dietary intake of cobalt and nail concentration. In this
study, a 2-fold risk of Barrett’s esophagus was also associated with higher toenail concentrations
of zinc.
The major limitation of this study, similar to the Rogers et al. study, however, is the exposure
assessment method, which is an analysis of a single sample of toenail clippings collected near the
time of diagnosis, with no accompanying data on potential sources of cobalt from the
environment or occupational exposure. In addition, reverse causality is possible due to trace
element deposition in nails being influenced by factors associated with cancer that could create a
bias away from the null, and explain the elevated cobalt levels found in this study. Barrett’s
esophagus, like esophageal cancer, has a prolonged latency period; however, it is unknown
whether the disease progression of Barrett’s esophagus could influence deposition of trace metals
in nails. Similar to the Rogers et al. study, co-exposures from other metals were not reported or
considered in the analysis of cobalt, and no correlations among the metals were reported.
Integration of the evidence across studies
While these two well-conducted population-based case-control studies in Ireland and in Western
Washington state reported relatively consistent findings, had adequate numbers of participants,
used sound methodologies, and demonstrated exposure-response relationships, the key issue of
temporality remains unaddressed. The dependence of these studies upon a single sample of
toenails collected at the time of diagnosis meant that neither had complete or even adequate data
on cobalt during the relevant windows of exposure throughout the natural history of the two
conditions to definitely establish temporality.
4.4
4.4.1
Cancer assessment: Other cancers
Other aerodigestive cancers - oral cavity, pharyngeal, and laryngeal cancers
The available data to evaluate cobalt in relation to other aerodigestive cancers, specifically
cancers of the oral cavity, pharynx, and larynx, consist of the electrochemical workers cohort
study (Mur et al. 1987), and one population-based case-control biomarker study (Rogers et al.
1993). The first publication from the electrochemical workers cohort (Mur et al.) provided an
SMR for buccal cavity, pharyngeal, and laryngeal cancers for those working in cobalt
production. Rogers et al. provided OR estimates of cobalt in toenails among incident laryngeal
cancers and oral cavity cancers and controls, and included exposure-response data as well. These
are rare cancers (incidence 11.0 per 100,000 men and women for oral cavity cancer; and 3.3 per
100,000 men and women for laryngeal cancers) (SEER 2015c); and unlike lung and esophageal
cancers, 5-year survival rates are much higher for oral cavity/pharyngeal and laryngeal cancers
(62.7% and 60.0%, respectively), suggesting that mortality statistics are less useful for informing
the cobalt and cancer assessment. Potential risk factors for these cancers include smoking and
other tobacco use, alcohol (tobacco and alcohol together are worse than either alone), asbestos,
and nickel.
The risk of death from buccal cavity, pharyngeal, and laryngeal cancer among electrochemical
workers was SMR = 3.36 (95% CI = 0.29 to 10.29), based on 2 observed deaths (Mur et al.
1987).
This draft document should not be construed to represent final NTP determination or policy
59
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Rogers et al. (1993) reported a borderline significantly elevated odds ratio for oral cavity cancer
for the highest level (≥ 0.17 ppm) of cobalt concentration in toenails compared to the lowest
level (< 0.05 ppm) of cobalt (OR = 1.9, 95% CI = 1.0 to 3.6). The OR was elevated but not
significant for those with medium levels (0.05 to 0.17 ppm) of cobalt concentration compared to
those with low levels (OR = 1.5, 95% CI = 0.9 to 2.6). The exposure-response test for trend was
not significant (P-value not reported). The finding was present in both in situ/localized tumors
and individuals with regional/distant tumors. In this study, diet was not found to be an
explanation for the higher risks, and tobacco and alcohol levels were controlled in the analyses.
A borderline significantly elevated odds ratio for laryngeal cancer was reported for medium
toenail levels (0.05 to 0.17 ppm) compared to the lowest level (< 0.05 ppm) of cobalt (OR = 2.0,
95% CI = 1.0 to 3.8). However, the OR for the highest level of cobalt was 1.0 (95% CI = 0.4 to
2.6), with no indication of a trend in exposure response.
As with esophageal cancer, it is not possible to assess the actual exposure levels among cases
and controls as they are combined at each concentration level. Because nails were collected after
diagnosis, to address potential reverse causation, cases were stratified by stage at diagnosis (in
situ/localized versus regional/distant) and by time from diagnosis to interview (< 7 months vs. ≥
7 months). No statistically significant differences in the odds ratios by time from diagnosis to
interview or stage of disease were observed, which argues against reverse causation.
With respect to these aerodigestive cancers, information is inadequate to evaluate the association
with exposure to cobalt based on findings from these two studies, one of which was
underpowered (Mur et al. 1987) and one of which had critical concerns regarding exposure
misclassification due to the use of a single sample of toenails collected at the time of diagnosis,
which might not have been the relevant window of exposure (Rogers et al. 1993).
4.4.2
Other cancers
The available data to evaluate cobalt in relation to other cancers is inadequate as it was primarily
limited to one cohort study reporting on multiple cancers (Tüchsen et al. 1996) and two studies
reporting on brain cancer (Tüchsen et al. 1996, Moulin et al. 1993) (data not shown). Neither of
the two studies had adequate numbers of exposed cases (2 cases or fewer) to evaluate brain
cancer risk from exposure to cobalt. Among porcelain painters exposed to cobalt dyes, the
authors reported that cervical cancer was elevated (SIR = 2.31, lower confidence limit > 1.0)
based on 12 exposed cases (Tüchsen et al. 1996). For other cancer sites with at least four cases,
elevated SIRs (not statistically significant) were also observed for ovary and other skin, and the
SIR was close to 1.0 for breast cancer.
4.5
Preliminary listing recommendation
There is inadequate evidence from studies in humans to evaluate the association between
exposure to cobalt and cancer. While almost all the cohort studies reported approximately a
doubling of the risk of lung cancer from exposure to various cobalt compounds, cobalt exposure
was likely correlated with exposure to other known lung carcinogens, which complicates the
interpretation of the results. Increased risks of esophageal cancer were found in two populationbased case-control studies; however, cobalt exposure was assessed in toenail samples at or after
60
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
05/07/15
cancer diagnosis. Thus, it is unclear whether cobalt levels in the toenails reflected exposure to
cobalt preceding cancer or resulted from changes due to tumor formation.
This draft document should not be construed to represent final NTP determination or policy
61
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
This Page Intentionally Left Blank
62
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
5 Studies of Cancer in Experimental Animals
This section reviews and assesses the evidence from carcinogenicity studies in experimental
animals exposed to cobalt and certain cobalt compounds. Cancer and co-carcinogen studies in
experimental animals were identified using methods described in the protocol and literature
search strategy document (http://ntp.niehs.nih.gov/go/730697). In all, 23 studies (16
carcinogenicity and 9 co-carcinogenicity studies) were identified that met the inclusion criteria.
Some of these publications overlap since some co-carcinogenicity studies had a cobalt exposure
alone group and a corresponding control as part of their design. The criteria to evaluate exposure
specific to cobalt and/or cobalt compounds require studies that either had observational durations
> 12 months for rats and mice, or were co-carcinogen exposure studies (initiation/promotion and
other co-carcinogen studies that isolate the effect of cobalt compound exposures) and that report
on the presence or absence of neoplastic and related non-neoplastic lesions. Several studies were
excluded from the review because they did not have concurrent controls. These included Hopps
et al. (1954), Delahant (1955), Gilman (1962), Nowak (1966), and Gunn et al. (1967). Studies of
cobalt alloys and radioactive cobalt in experimental animals were not considered to be
informative because of potential confounding by other carcinogens. (See IARC 2006 for a
review of studies of cobalt alloys.)
This section is organized by the type of study, i.e., carcinogenicity (Section 5.1) and
co-carcinogenicity (Section 5.2). For each of these study types, the monograph provides an
overview of the available studies, assesses their quality, discusses the findings and identifies
potential treatment-related cancer sites (carcinogenicity studies only). The co-carcinogen studies
are only briefly discussed because they do not contribute substantially to the evaluation of
potential carcinogenicity. Section 5.3 provides a synthesis of the findings for the different types
of cobalt compounds across the cancer sites. The preliminary level of evidence conclusion for
the carcinogenicity of cobalt and certain cobalt compounds as a class from studies in
experimental animals is provided in Section 7, which provides the rationale for evaluating them
as a class.
5.1
5.1.1
Carcinogenicity studies
Overview of the studies
Different forms of cobalt were tested in 16 carcinogenicity studies: cobalt metal or cobalt
nanoparticles (6 studies); two soluble cobalt salts, cobalt sulfate heptahydrate (2 studies) and
cobalt chloride (1 study); and two poorly soluble cobalt compounds, cobalt oxide (6 studies) and
cobalt sulfide (1 study); (see Table 5-1). Most carcinogenicity studies were conducted in rats,
with three studies in mice, and one study in hamsters. Routes of administration included either
administration through the respiratory tract (inhalation or intratracheal instillation) or by local
injection (subcutaneous, intramuscular, intraperitoneal, intrapleural, or intrarenal).
This draft document should not be construed to represent final NTP determination or policy
63
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 5-1. Overview of cancer studies in experimental animals reviewed
Strain (sex)
Substance
Route
Exposure period/
study duration
Cobalt metal
Inhalation
2 yr/2 yr
(NTP 2014b)
Cobalt metal
Inhalation
2 yr/2 yr
(NTP 2014b)
Cobalt metal [nano] and
IM inj.
Single dose/
1 yr
Reference
Cobalt metal
Rat F344/NTac (M&F)
Mouse B6C3F1 (M&F)
Rat Sprague-Dawley (M)
(Hansen et al. 2006)
Cobalt metal [bulk]
SC inj.
Rat Sprague-Dawley (F)
Cobalt metal
Intrarenal inj.
Rat Hooded (F)
Cobalt metal
Intrapleural inj.
Rat Hooded (M&F)
Cobalt metal
IM inj.
Single dose/
1 yr
Single dose/
2.3 yr
Single dose/lifespan
Inhalation
2 yr/2 yr
(NTP 1998)
Inhalation
2 yr/2 yr
(NTP 1998)
Rat Wistar (M)
Cobalt sulfate
heptahydrate
Cobalt sulfate
heptahydrate
Cobalt chloride
SC inj.
8–12 mo/8–12 mo
Poorly soluble cobalt compounds
Rat Sprague-Dawley (M&F)
Cobalt(II) oxide
Intratracheal instill.
1.5 yr/lifespan
(Steinhoff and Mohr 1991)
Rat Sprague-Dawley (M&F)
Cobalt(II) oxide
IP inj.
6 mo/lifespan
(Steinhoff and Mohr 1991)
Rat Sprague-Dawley (M)
Cobalt(II) oxide
SC inj.
730 day/lifespan
(Steinhoff and Mohr 1991)
Rat Wistar (M&F)
Cobalt(II)oxide
IM inj.
Single dose/1.3 yr
(Gilman and Ruckerbauer
1962)
Mouse Swiss (F)
Cobalt(II) oxide
IM inj.
Single dose/2 yr
(Gilman and Ruckerbauer
1962)
Hamster Syrian Golden (M)
Cobalt(II) oxide
Inhalation
Lifespan/lifespan
(Wehner et al. 1977)
Cobalt sulfide
Intrarenal inj.
Single dose/1 yr
(Jasmin and Riopelle 1976)
Soluble cobalt compounds
Rat F344/N (M&F)
Mouse B6C3F1 (M&F)
Rat Sprague-Dawley (F)
M = male, F = female, instill. = instillation, inj. = injection, IP = intraperitoneal, IM = intramuscular, SC = subcutaneous, wk = week, yr = year.
64
This draft document should not be construed to represent final NTP determination or policy
(Jasmin and Riopelle 1976)
(Heath and Daniel 1962)
(Heath 1956)
(Shabaan et al. 1977)
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
5.1.2
06/05/15
Study quality assessment
Each of these primary studies was systematically evaluated for its ability to inform the cancer
hazard evaluation using a series of signaling questions related to the following study
performance elements: population, exposure conditions, outcome assessment, potential
confounding, and statistics and reporting (see Protocol for Preparing the RoC Monograph on
Cobalt [NTP 2014c]). An overview of the quality evaluations for the carcinogenicity studies is
shown in Table 5-3 and discussed below. Details of each study assessment and quality criteria on
a study-by-study basis are reported in Appendix D.
No critical concerns for biases were identified in any of the 16 carcinogenicity studies and they
were all considered to have some utility for the cancer hazard evaluation. The four NTP
inhalation studies (cobalt metal and cobalt sulfate in rats and mice) were considered to be the
most informative (high utility) because they used a sufficient number of experimental animals of
both sexes for a near lifetime exposure duration and tested three dose levels along with an
untreated control. Two inhalation/intratracheal instillation studies of exposure to cobalt oxide
(Steinhoff and Mohr 1991, Wehner et al. 1977) and three injection studies of cobalt metal or
cobalt sulfide in two publications (Hansen et al. 2006, Steinhoff and Mohr 1991) were
considered to have moderate utility. In general, most of the limitations of the studies were related
to low sensitivity of the study to detect an effect, e.g., due to the use of a single dose, short study
duration, or small numbers of animals. In the remaining seven injection studies (Heath 1956,
Heath and Daniel 1962, Gilman and Ruckerbauer 1962, Jasmin and Riopelle 1976, Shabaan et
al. 1977), there were major concerns for several potential biases; thus, these studies were
considered to have lower utility. Most of these studies had low sensitivity or incomplete
necropsies. Poor reporting of methods and results was also a common problem and in some
studies there were concerns about potential confounding. Historical controls from a related study
by the same authors were used in lieu of concurrent controls in one study (Heath and Daniel
1962). Overall, the major limitations in the studies with low and moderate utility were primarily
(but not exclusively) due to low sensitivity and for these cases there is little concern that these
limitations would decrease confidence in a positive finding.
This draft document should not be construed to represent final NTP determination or policy
65
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 5-2. Overview of experimental animal carcinogenicity study quality evaluations
Duration
Reporting
& analysis
Confounding
Pathology
Overall
utility
Cobalt
metal
+++
Yes
+++
+++
+++
++
+++
+++
+++
+++
+++
+++
High
(NTP
2014b) -M
Cobalt
metal
+++
Yes
+++
+++
+++
++
+++
+++
+++
+++
+++
+++
High
(Hansen et
al. 2006)a
Cobalt
metal
and nano
+++
No
NR
NR
++
+++
+++
++
++
++
+
+
Moderate
(Jasmin
and
Riopelle
1976)a
Cobalt
metal
and
sulfide
+++
No
NR
++
+
NR
++
++
++
++
++
+
Low
(Heath
and
Daniel
1962)
Cobalt
metal
+
Yesb
NR
++
+
NR
++
++
+
++
+
+++
Low
(Heath
1956)
Cobalt
metal
++
Yes b
NR
++
+
NR
++
++
+
++
+
+++
Low
(NTP
1998)- R
Cobalt
sulfate
+++
Yes
+++
+++
+++
+++
+++
+++
+++
+++
+++
+++
High
(NTP
1998)- M
Cobalt
sulfate
+++
Yes
+++
+++
+++
+++
+++
+++
+++
+++
+++
+++
High
(Shabaan
et al. 1977)
Cobalt
chloride
++
Yes b
NR
NR
+
++
+
+
+
++
++
+
Low
66
This draft document should not be construed to represent final NTP determination or policy
Stat
power
(NTP
2014b) -R
Animal
model
Dosing
Treatmentrelated
survival
Sensitivity
Purity
Randomization
Historical
data
Quality
Controls
Study
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Overall
utility
Duration
Stat
power
Animal
model
Reporting
& analysis
Confounding
Pathology
Sensitivity
Treatmentrelated
survival
Dosing
Purity
Randomization
Historical
data
Quality
Controls
Study
06/05/15
(Steinhoff
and Mohr
1991)(intratrac
heal)
Cobalt
oxide
+++
No
NR
++
++
++
++
++
++
+++
+++
+++
Moderate
(Steinhoff
and Mohr
1991) (IP)
Cobalt
oxide
+++
No
NR
++
+
NR
++
++
++
+++
+
+++
Moderate
(Steinhoff
and Mohr
1991) (SC)
Cobalt
oxide
+++
No
NR
++
++
NR
++
++
++
++
+
+++
Moderate
(Gilman
and
Ruckerba
uer 1962)
–R
Cobalt
oxide
+++
No
NR
+
+
++
++
+
+
+++
+
++
Low
(Gilman
and
Ruckerba
uer 1962)
–M
Cobalt
oxide
+++
No
NR
+
+
++
++
+
++
++
++
+++
Low
(Wehner
et al. 1977)
Cobalt
oxide
+++
No
NR
++
++
+++
+++
++
+
++
+
+++
Moderate
+++ = high quality/little to no concerns, ++ = moderate quality/moderate concerns, + = low quality/high concerns, 0 = inadequate, NR = not reported; M = mice; R = rats.
a
Includes test results for two forms of cobalt, so considered two studies.
b
Limited number of controls (less than 15) from an earlier study.
This draft document should not be construed to represent final NTP determination or policy
67
06/05/15
5.1.3
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Assessment of neoplastic findings from carcinogenicity studies
Discussions of the findings from the 16 carcinogenicity studies grouped by site of tumor
development are reported below and in Tables 5-3 to 5-5. The main neoplasm locations were the
lung in inhalation and intratracheal studies (six studies) and injection sites in studies using
various routes of injection (subcutaneous, intramuscular, intraperitoneal, intrarenal, and
intrapleural). In addition, in some inhalation studies, some tumors were observed in sites distal
from the route of administration. Findings for cobalt compounds across organ sites are discussed
in Section 5.3.
Lung (Table 5-3)
Different types of cobalt compounds – cobalt metal (NTP 2014b), a soluble cobalt salt, cobalt
sulfate heptahydrate (NTP 1998), and a poorly soluble cobalt compound, cobalt oxide (Steinhoff
and Mohr 1991) – caused lung neoplasms after exposure by inhalation or intratracheal
instillation. Study results for six respiratory exposure studies are reported in Table 5-3 including
two studies in mice, three studies in rats and one study in hamsters. Four of these studies were
high-quality, well-designed, and well-conducted studies (NTP 2014b, 1998) and all had either
high (NTP 2014b, 1998) or moderate (Steinhoff and Mohr 1991, Wehner et al. 1977) utility for
evaluating potential cancer hazards.
Four studies found strong evidence that cobalt (both cobalt metal and cobalt sulfate) causes lung
tumors in both mice and rats (NTP 2014b, 1998). Significant dose-related increases were seen
for alveolar/bronchiolar carcinoma and for alveolar/bronchiolar adenoma or carcinoma combined
in all dose groups (low, 1.25 mg/m3; medium, 2.5 mg/m3; high, 5 mg/m3) in male and female
mice and rats exposed to cobalt metal by inhalation (NTP 2014b). The incidences of
alveolar/bronchiolar adenoma were also significantly increased in rats and mice, although not
always in all dose groups. The incidences of carcinoma was very high; when adjusted for
intercurrent mortality, incidences in the high-dose groups were 81% for male rats, 69% for
female rats, 94% for male mice, and 88% for female mice. In addition, dose-related significant
increases in multiplicity (animals with more than one lung tumor) of carcinoma were also found
for all dose groups in male and female mice and male rats and in the high-dose (5 mg/m3) groups
for female rats (NTP 2014b). Female rats also had, in all dose groups, non-significant increases
in cystic keratinizing epithelioma, which is a benign squamous cell neoplasm that can progress to
squamous-cell carcinoma. Cystic keratinizing epithelioma (CKE) is considered to be exposure
related in females, because it is very rare and a single squamous-cell carcinoma was also
observed in the high dose group. In males, a single CKE was found in each of the low and high
exposure groups, and may have been exposure related. Lesions of alveolar or bronchiolar
epithelial hyperplasia, which can progress to neoplasms, was also significantly increased in both
sexes of rats and mice in all dose levels tested, except for bronchiolar epithelium hyperplasia in
mice, which were significantly increased at mid- and high-dose groups in females and high-dose
group in males.
In the NTP (1998) inhalation studies of cobalt sulfate heptahydrate, significant dose-related
increases were observed for alveolar/bronchiolar carcinoma, and alveolar/bronchiolar adenoma
in male and female mice (high dose, 3.0 mg/m3) and female rats (high and mid dose, 1.0 mg/m3)
and for alveolar/bronchiolar carcinoma or adenoma combined for male rats (high dose) (NTP
1998). A single squamous-cell carcinoma was also found in the mid- and high-dose groups of
68
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
female rats. Non-neoplastic lesions of alveolar or bronchiolar epithelial hyperplasia (considered
pre-neoplastic) and metaplasia were also significantly increased in both sexes of rats, but not in
mice.
The fifth study reported significant increases in lung neoplasms (alveolar/bronchiolar adenoma,
benign squamous epithelial neoplasm, or alveolar/bronchiolar carcinoma combined) in male rats
administered cobalt oxide by intratracheal instillation (Steinhoff and Mohr 1991). Nonsignificant increases in lung neoplasms (alveolar/bronchiolar carcinoma and alveolar/bronchiolar
adenoma) were seen in females. There were significant increases in alveolar/bronchiolar
proliferation (types of lesions not described) in both sexes combined. Histological examinations
were performed on all high-dose group animals; in the low-dose group, only those with gross
lesions were examined, which could underestimate the incidence by not detecting microscopic
neoplasms.
In the last study, lung tumors were not observed in hamsters exposed to cobalt oxide by
inhalation, although exposure did cause pneumoconiosis, which was evidenced by a variety of
lesions including, e.g., interstitial pneumonitis, diffuse granulomatous pneumonia, fibrosis of
alveolar septa, and bronchial and bronchiolar epithelial (basal cell) hyperplasia (Wehner et al.
1977). There was relatively poor survival among the cobalt-treated animals and the
corresponding dust sham-treated controls, which may have limited the sensitivity to detect an
effect. In addition, hamsters have been described as a less sensitive model for detecting lung
tumors than rats or mice (McInnes et al. 2013, Steinhoff and Mohr 1991). (Findings not reported
in Table 5-3 because no tumors were observed.)
This draft document should not be construed to represent final NTP determination or policy
69
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 5-3. Lung neoplasms and non-neoplastic lesions in experimental animals exposed to cobalt compounds
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
NTP 2014b
Rat (F344/NTac)
Male
105 wk
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Multiple alveolar/bronchiolar carcinoma
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
6/50 (12%)*
2.5 mg/m³
16
14/50 (28%)**
5 mg/m³
16
30/50 (60%)**
Alveolar/bronchiolar carcinomaa
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
16/50 (38%)***
2.5 mg/m³
16
34/50 (77%)***
5 mg/m³
16
36/50 (81%)***
Trend-test P-value: 0.001
Multiple alveolar/bronchiolar adenoma
0 mg/m³
17
1/50 (2%)
1.25 mg/m³
20
3/50 (6%)
2.5 mg/m³
16
2/50 (4%)
5 mg/m³
16
6/50 (12%)
Alveolar/bronchiolar adenomaa
0 mg/m³
17
2/50 (5%)
1.25 mg/m³
20
10/50 (24%)*
2.5 mg/m³
16
10/50 (23%)*
5 mg/m³
16
14/50 (33%)***
Trend-test P-value: 0.011
Alveolar/bronchiolar carcinoma or adenoma
combineda
0 mg/m³
70
17
2/50 (5%)
This draft document should not be construed to represent final NTP determination or policy
Comments, strengths,
and limitations
Survival in exposed
groups was similar to
controls.
Strengths: A welldesigned study in all
factors.
Limitations:
Decreases in body
weight in mid and high
dose rats,
Other comments:
Historical controls
were limited, as
Fischer 344/NTac rats
have only been used in
two NTP
carcinogenicity studies
and so it is based on
only 100 rats.
Significantly increased
non-neoplastic lesions:
Alveolar epithelium
hyperplasia (preneoplastic) - all dose
levels
Bronchiolar
hyperplasia (preneoplastic) - all dose
levels
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
06/05/15
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
1.25 mg/m³
20
25/50 (58%)***
2.5 mg/m³
16
39/50 (85%)***
5 mg/m³
16
44/50 (94%)***
Comments, strengths,
and limitations
Trend-test P-value: 0.001
Cystic keratinizing epithelioma
NTP 2014b
Rat (F344/NTac)
Female
105 wk
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
1/50 (2%)
2.5 mg/m³
16
0/50 (0%)
5 mg/m³
16
1/50 (2%)
Multiple alveolar/bronchiolar carcinoma
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
4/50 (8%)
2.5 mg/m³
24
3/50 (6%)
5 mg/m³
25
18/50 (36%)**
Alveolar/bronchiolar carcinomaa
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
9/50 (21%)***
2.5 mg/m³
24
17/50 (42%)***
5 mg/m³
25
30/50 (69%)***
Trend-test P-value: 0.001
Multiple alveolar/bronchiolar adenoma
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
1/50 (2%)
2.5 mg/m³
24
3/50 (6%)
5 mg/m³
25
4/50 (8%)
Alveolar/bronchiolar adenomaa
This draft document should not be construed to represent final NTP determination or policy
Survival was
significantly decreased
in the mid-dose group.
Strengths: A welldesigned study in
almost all factors.
Limitations: A
significant decrease in
survival of female rats
and decreases in body
weight in mid- and
high-dose rats.
Other comments:
Historical controls
were limited, as
Fischer 344/NTac rats
have only been used in
two NTP
carcinogenicity studies
and so it is based on
only 100 rats.
Significantly increased
71
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
0 mg/m³
35
2/50 (5%)
1.25 mg/m³
26
7/50 (16%)
2.5 mg/m³
24
9/50 (22%)*
5 mg/m³
25
13/50 (31%)**
Dose levels
Trend-test P-value: 0.002
Alveolar/bronchiolar carcinoma or adenoma
combineda
0 mg/m³
35
2/50 (4%)
1.25 mg/m³
26
15/50 (35%)***
2.5 mg/m³
24
20/50 (49%)***
5 mg/m³
25
38/50 (86%)***
Comments, strengths,
and limitations
non-neoplastic lesions:
Alveolar hyperplasia
(pre-neoplastic) - all
dose levels.
Bronchiolar
hyperplasia (preneoplastic) - all dose
levels
Trend-test P-value: 0.001
Squamous cell carcinoma
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
0/50 (0%)
2.5 mg/m³
24
0/50 (0%)
5 mg/m³
25
1/50 (2%)
Cystic keratinizing epitheliomaa
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
4/50 (10%)i
2.5 mg/m³
24
1/50 (3%)i
5 mg/m³
25
2/50 (5%)i
Trend-test P-value: 0.002
NTP 2014b
Mouse (B6C3F1/N)
Male
105 wk
72
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Multiple alveolar/bronchiolar carcinoma
0 mg/m³
39
3/50 (6%)
1.25 mg/m³
31
18/49 (36%)**
2.5 mg/m³
29
24/50 (48%)**
This draft document should not be construed to represent final NTP determination or policy
Survival significantly
decreased at 2.5 and 5
mg/m³.
Strengths: A welldesigned study in
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
5 mg/m³
06/05/15
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
25
36/50 (72%)**
almost all factors.
Limitations: A
significant decrease in
survival of male mice
and decrease in body
weight in high dose
mice.
Other comments:
Significantly increased
non-neoplastic lesions:
Alveolar/bronchiolar
epithelium hyperplasia
(pre-neoplastic) - all
dose levels
Alveolar epithelium
hyperplasia (preneoplastic) - all dose
levels
Bronchiolar epithelium
hyperplasia (preneoplastic) - high dose
Trend-test P-value: 0.001
Alveolar/bronchiolar carcinomaa
0 mg/m³
39
11/50 (23%)
1.25 mg/m³
31
38/49 (79%)***
2.5 mg/m³
29
42/50 (88%)***
5 mg/m³
25
46/50 (94%)***
Trend-test P-value: 0.001
Multiple alveolar/bronchiolar adenoma
0 mg/m³
39
0/50 (0%)
1.25 mg/m³
31
1/49 (2%)
2.5 mg/m³
29
1/50 (2%)
5 mg/m³
25
0/50 (0%)
Alveolar/bronchiolar adenomaa
0 mg/m³
39
7/50 (15%)
1.25 mg/m³
31
11/49 (25%)
2.5 mg/m³
29
15/50 (36%)*
5 mg/m³
25
3/50 (7%)
Alveolar/bronchiolar carcinoma or adenoma
combineda
0 mg/m³
39
16/50 (33%)
1.25 mg/m³
31
41/49 (85%)***
2.5 mg/m³
29
43/50 (90%)***
5 mg/m³
25
47/50 (96%)***
Trend-test P-value: 0.001
NTP 2014b
Mouse (B6C3F1/N)
Cobalt metal
(98% pure, mass
Inhalation
6 hr/day, 5 day/wk ×
Multiple alveolar/bronchiolar carcinoma
0 mg/m³
36
1/49 (10%)
This draft document should not be construed to represent final NTP determination or policy
Survival in exposed
groups was similar to
73
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Female
105 wk
median aerodynamic
diameter 1-3 μm)
105 wk
1.25 mg/m³
36
7/50 (50%)*
2.5 mg/m³
27
20/50 (76%)**
5 mg/m³
26
24/50 (86%)**
Trend-test P-value: 0.001
Alveolar/bronchiolar carcinomaa
0 mg/m³
36
5/49 (11%)
1.25 mg/m³
36
25/50 (54%)***
2.5 mg/m³
27
38/50 (79%)***
5 mg/m³
26
43/50 (88%)***
Trend-test P-value: 0.001
Multiple alveolar/bronchiolar adenoma
0 mg/m³
36
0/49 (0%)
1.25 mg/m³
36
1/50 (2%)
2.5 mg/m³
27
0/50 (0%)
5 mg/m³
26
1/50 (2%)
Alveolar/bronchiolar adenomaa
0 mg/m³
36
3/49 (7%)
1.25 mg/m³
36
9/50 (20%)
2.5 mg/m³
27
8/50 (19%)
5 mg/m³
26
10/50 (25%)*
Trend-test P-value: 0.037
Alveolar/bronchiolar carcinoma or adenoma
combineda
74
0 mg/m³
36
8/49 (18%)
1.25 mg/m³
36
30/50 (64%)***
2.5 mg/m³
27
41/50 (85%)***
5 mg/m³
26
45/50 (92%)***
This draft document should not be construed to represent final NTP determination or policy
Comments, strengths,
and limitations
controls.
Strengths: A welldesigned study in all
factors.
Limitations: Decrease
in body weight in high
dose mice.
Other comments:
Significantly increased
non-neoplastic lesions:
Alveolar/bronchiolar
epithelium hyperplasia
(pre-neoplastic) - all
dose levels;
Alveolar epithelium
hyperplasia (preneoplastic) - all dose
levels;
Bronchiolar epithelium
hyperplasia (preneoplastic) – mid- and
high-dose levels
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Trend-test P-value: 0.001
NTP 1998
Rat (F344)
Male
2 yr
Cobalt sulfate
(99% pure)
Inhalation
6 hr/day, 5 days/wk ×
105 wk
Alveolar/bronchiolar carcinomab
0 mg/m³
17
0/50 (0%)
0.3 mg/m³
15
0/50 (0%)
1.0 mg/m³
21
3/48 (11%)
3.0 mg/m³
15
1/50 (7%)
Alveolar/bronchiolar adenomab
0 mg/m³
17
1/50 (2%)
0.3 mg/m³
15
4/50 (18%)
1.0 mg/m³
21
1/48 (2%)
3.0 mg/m³
15
6/50 (28%)
Alveolar/bronchiolar adenoma or carcinoma
combinedb
0 mg/m³
17
1/50 (2%)
0.3 mg/m³
15
4/50 (18%)
1.0 mg/m³
21
4/48 (13%)
3.0 mg/m³
15
7/50 (34%)*
Survival in exposed
groups was similar to
controls.
Strengths: A welldesigned study in all
factors and survival
was similar to
controls..
Limitations: None.
Other comments:
Significantly increased
non-neoplastic lesions:
Alveolar epithelium
metaplasia - all dose
levels;
Alveolar epithelium
hyperplasia (preneoplastic) - all dose
levels
Trend-test P-value: 0.032
NTP 1998
Rat (F344)
Female
2 yr
Cobalt sulfate
(99% pure)
Inhalation
6 hr/day, 5 days/wk ×
105 wk
Alveolar/bronchiolar carcinomab
0 mg/m³
28
0/50 (0%)
0.3 mg/m³
25
2/49 (8%)
1.0 mg/m³
26
6/50 (20%)*
3.0 mg/m³
30
6/50 (18%)*
Trend-test P-value: 0.023
Alveolar/bronchiolar adenomab
0 mg/m³
28
0/50 (0%)
0.3 mg/m³
25
1/49 (3%)
This draft document should not be construed to represent final NTP determination or policy
Survival in exposed
groups was similar to
controls.
Strengths: A welldesigned study in all
factors and survival
was similar to controls.
Limitations: None.
Other comments:
Significantly increased
non-neoplastic lesions:
Alveolar epithelium
75
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
1.0 mg/m³
26
10/50 (36%)***
3.0 mg/m³
30
9/50 (30%)***
metaplasia - all dose
levels;
Alveolar epithelium
hyperplasia (preneoplastic) - high dose;
Alveolar epithelium
hyperplasia, atypical
(pre-neoplastic) - high
dose
Trend-test P-value: 0.001
Alveolar/bronchiolar adenoma or carcinoma
combinedb
0 mg/m³
28
0/50 (0%)
0.3 mg/m³
25
3/49 (11%)c
1.0 mg/m³
26
15/50 (51%)***c
3.0 mg/m³
30
15/50 (46%)***c
Trend-test P-value: 0.001
Squamous cell carcinoma
0 mg/m³
28
0/50 (0%)
0.3 mg/m³
25
0/49 (0%)
1.0 mg/m³
26
1/50 (2%)
3.0 mg/m³
30
1/50 (2%)
Alveolar/bronchiolar adenoma, carcinoma, or
squamous cell carcinoma combinedb
0 mg/m³
28
0/50 (0%)
0.3 mg/m³
25
3/49 (11%)
1.0 mg/m³
26
16/50 (54%)***
3.0 mg/m³
30
16/50 (49%)***
Trend-test P-value: 0.001
NTP 1998
Mice (B6C3F1)
Male
2 yr
76
Cobalt sulfate
(99% pure)
Inhalation
6 hr/day, 5 days/wk ×
105 wk
Alveolar/bronchiolar carcinomab
0 mg/m³
22
4/50 (13%)
0.3 mg/m³
31
5/50 (16%)
1.0 mg/m³
24
7/50 (25%)
3.0 mg/m³
20
11/50 (44%)*d
This draft document should not be construed to represent final NTP determination or policy
Survival in exposed
groups was similar to
controls.
Strengths: A welldesigned study in all
factors and survival
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Trend-test P-value: 0.006
Alveolar/bronchiolar adenoma
b
0 mg/m³
22
9/50 (30%)
0.3 mg/m³
31
12/50 (31%)
1.0 mg/m³
24
13/50 (41%)
3.0 mg/m³
20
18/50 (55%)*e
Comments, strengths,
and limitations
was similar to controls.
Limitations: None.
Trend-test P-value: 0.018
Alveolar/bronchiolar carcinoma or adenoma
combinedb
0 mg/m³
22
11/50 (36%)
0.3 mg/m³
31
14/50 (37%)
1.0 mg/m³
24
19/50 (57%)
3.0 mg/m³
20
28/50 (79%)***f
Trend-test P-value: 0.001
NTP 1998
Mice (B6C3F1)
Female
2 yr
Cobalt sulfate
(99% pure)
Inhalation
6 hr/day, 5 days/wk ×
105 wk
Alveolar/bronchiolar carcinomab
0 mg/m³
34
1/50 (3%)
0.3 mg/m³
37
1/50 (3%)
1.0 mg/m³
32
4/50 (9%)
3.0 mg/m³
28
9/50 (25%)**g
Trend-test P-value: 0.001
Survival in exposed
groups was similar to
controls.
Strengths: A welldesigned study in all
factors and survival
was similar to controls.
Limitations: None.
Alveolar/bronchiolar adenomab
0 mg/m³
34
3/50 (9%)
0.3 mg/m³
37
6/50 (15%)
1.0 mg/m³
32
9/50 (25%)
3.0 mg/m³
28
10/50 (33%)*h
Trend-test P-value: 0.024
Alveolar/bronchiolar carcinoma or adenoma
This draft document should not be construed to represent final NTP determination or policy
77
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
combinedb
0 mg/m³
34
4/50 (12%)
0.3 mg/m³
37
7/50 (18%)
1.0 mg/m³
32
13/50 (33%)*i
3.0 mg/m³
28
18/50 (50%)***i
Trend-test P-value: 0.001
Steinhoff and Mohr
1991
Rat (Sprague-Dawley)
Male
life-span
Cobalt oxide
("Chemically pure."
80% of particles were
5–40 μm)
Intratracheal instillation
1 dose/2 wk × 18 doses,
then 1 dose/4 weeks ×
11 doses (up to 30th
dose), then 1 dose/2
weeks × 9 doses (total
39 doses)
Bronchio/alveolar carcinoma
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
0/50 (0%)
10 mg/kg bw
NR
3/50 (6%) 1
Bronchio/alveolar adenoma
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
0/50 (0%)
10 mg/kg bw
NR
2/50 (4%)
Bronchioalveolar adenomas or
bronchioalveolar carcinomas combined
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
0/50 (0%)
10 mg/kg bw
NR
5/50 (10%)*
Benign squamous epithelial tumor
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
1/50 (2%)
10 mg/kg bw
NR
0/50 (0%)
1
78
This draft document should not be construed to represent final NTP determination or policy
Survival and body
weight were the same
as controls.
Strengths: Two dose
levels were tested in a
high number of both
sexes of rats for two
years, with
observations for the
lifespan without any
significant difference
in survival compared
to untreated controls.
Limitations: Only the
high-dose group
received full
necropsies. Details of
the chemical and
animal husbandry were
not reported.
Other comments:
Significantly increased
non-neoplastic lesions:
Bronchio/alveolar
proliferation - both
dose levels.
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Steinhoff and Mohr
1991
Rat (Sprague-Dawley)
Female
life-span
Substance & purity
Dosing regimen
Cobalt oxide
("Chemically pure",
80% of particles were
5–40 μm)
Intratracheal instillation
1 dose/2 wk × 18 doses,
then 1 dose/4 weeks ×
11 doses (up to 30th
dose), then 1 dose/2
weeks × 9 doses (total
39 doses)
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Bronchio/alveolar carcinoma
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
0/50 (0%)
10 mg/kg bw
NR
1/50 (2%)
Bronchio/alveolar adenoma
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
1/50 (2%)
10 mg/kg bw
NR
0/50 (0%)
Bronchio/alveolar adenoma or
bronchio/alveolar carcinoma combined
0 mg/kg bw
NR
0/50 (0%)
2 mg/kg bw
NR
1/50 (2%)
10 mg/kg bw
NR
1/50 (2%)
Comments, strengths,
and limitations
Survival in exposed
groups was similar to
controls.
Strengths: Two dose
levels were tested in a
high number of both
sexes of rats for two
years, with
observations for the
lifespan without any
significant difference
in survival compared
to untreated controls.
Limitations: Only the
high-dose group
received full
necropsies. Details of
the chemical and
animal husbandry were
not reported.
Other comments:
Significantly increased
non-neoplastic lesions:
Bronchio/alveolar
proliferation - both
dose levels.
* = P-value ≤ 0.05; ** = P-value ≤ 0.01; *** = P-value ≤ 0.001. NR = Not reported.
+
= Number of animals necropsied for NTP 2014b and NTP 1998 and is the number of animals at the beginning of the study for all other studies.
a
Adjusted percent incidence based on Poly-3 estimated neoplasm incidence after adjustment for intercurrent mortality.
b
Adjusted percent incidence based on Kaplan-Meier estimated incidence at the end of the study after adjustment for intercurrent mortality.
c
Increased over historical control levels with a mean of 7/650 and range of 0% to 4%.
d
Increased over historical control levels with a mean of 75/947 and range of 0% to 16%.
e
Increased over historical control levels with a mean of 141/947 and range of 6% to 36%.
f
Increased over historical control levels with a mean of 205/947 and range of 10% to 42%.
g
Increased over historical control levels with a mean of 38/939 and range of 0% to 12%.
h
Increased over historical control levels with a mean of 61/939 and range of 0% to 14%.
This draft document should not be construed to represent final NTP determination or policy
79
06/05/15
i
j
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Increased over historical control levels with a mean of 97/939 and range of 0% to 16%.
Includes adenocarcinoma (2) and bronchioalveolar adenocarcinoma (1).
80
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Injection sites (subcutaneous, intramuscular, intraperitoneal, intrapleural, and intrarenal)
Exposure to several different cobalt forms (cobalt metal, cobalt chloride, and cobalt oxide) by
injection increased injection-site tumors in several studies in rats (Hansen et al. 2006, Steinhoff
and Mohr 1991, Shabaan et al. 1977, Gilman and Ruckerbauer 1962, Heath and Daniel 1962,
Heath 1956). However, no injection tumors were observed in other studies in rats (Hansen et al.
2006, Jasmin and Riopelle 1976) or in the only study in mice (Gilman and Ruckerbauer 1962).
Differences in dose levels, sex, and inadequate statistical power could explain these different
findings. These studies were considered to have moderate (Steinhoff and Mohr 1991, Hansen et
al. 2006) or low utility (Heath et al. 1956, Heath and Daniel 1962, Gilman and Ruckerbauer
1962, Jasmin and Riopelle 1976, Shabaan et al. 1977). However, many concerns for potential
biases were related to sensitivity such as limited dosing regimens and statistical power and thus
would not necessarily decrease confidence in positive findings. Many studies also had limited
reporting, which in part may be typical of older studies (published in the 1950s to 1970s). The
relevance of injection studies for evaluating carcinogenicity in humans is discussed in the
synthesis (Section 4.4).
Injection of cobalt metal (nanoparticles or microparticles) caused significant increases in the
incidences of various types of sarcoma in several studies. Hansen et al. (2006) directly compared
potential carcinogenic effects of cobalt metal nanoparticles and larger size cobalt metal particles
in rats. However, both sizes of particles were placed into the same animals; cobalt nanoparticles
were administered intramuscularly and bulk cobalt metal was administered subcutaneously. The
study also used a similar design to test other materials (nickel, titanium dioxide, and silicon
dioxide). Cobalt-treated animals were sacrificed at 6 and 8 months (due to mortality from
tumors) and compared to controls, which were administered PVC and sacrificed at six and
twelve months. Local sarcomas developed around the site of the nanoparticles in one of four rats
at the 6-month sacrifice and in 5 of 6 rats at the 8-month sacrifice. No tumors were observed
around the injection site of the bulk cobalt metal at either sacrifice time, although a single lesion
of local fibroblastic proliferation occurred in one of six rats sacrificed at 8 months. The short
duration period of 8 months limited the ability to see if the fibroblastic proliferation cause by
microparticles would progress into neoplasms. The study also had limited statistical power
because of small numbers of animals in the exposed and control groups. With respect to the other
materials, tumors were observed in animals after implantation (nanoparticles) or subcutaneous
injection (bulk) with nickel but not with injections of titanium dioxide or silicon dioxide. The
ratio of surface area to volume between the Ni/Co and other compounds was not significantly
different, which suggests that the neoplasms were not mediated by physical events and thus
supports that the carcinogenic effect is due to cobalt.
A series of studies in hooded rats (Heath and Daniel 1962, Heath 1956) that injected cobalt metal
by different exposure routes rats reported sarcomas – rhabdomyofibrosarcoma (including in the
heart, intercostal muscle), rhabdomyosarcoma, fibrosarcoma, or other sarcoma – at the site of
injection, but not in the controls. The earlier study (Heath 1956) injected cobalt into male and
female rats intramuscularly in the thigh and the later study injected cobalt into the intrathoracic
region (Heath and Daniel 1962). The controls from the 1956 study were used for the 1962 study.
Rhabdomyofibrosarcoma, especially cardiac rhabdomyofibrosarcoma, are very rare tumors.
Evidence that the sarcomas were caused by a local carcinogenic effect, beyond the fact that they
only developed at injection sites, was seen by their tissue of origin. The 1962 study was limited
This draft document should not be construed to represent final NTP determination or policy
81
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
by poor survival at the beginning of the study (8 rats died within three days) caused by the
injections. Sarcomas originating from muscle tissue were only found in studies that injected
cobalt metal by intramuscular injection (rhabdomyofibrosarcoma or rhabdomyosarcoma) or
intrapleural injection (cardiac or intercostal muscle rhabdomyosarcoma). Relatively high
incidences in sarcomas were observed in both studies although the studies had limited sensitivity
because only a few animals were tested at only one dose.
In contrast, no neoplasms were reported in a study in which cobalt metal was injected directly
into the kidney of female rats, (Jasmin and Riopelle 1976). Compared to the other injection site
studies that used a single dose, Jasmin and Riopelle used a lower dose (10 mg/rat) than those
used in the studies that induced neoplasms (> 20 mg/rat) (Gilman and Ruckerbauer 1962, Heath
and Daniel 1962, Heath 1956), suggesting that the dose might have been to low; in addition the
study duration was only 12 months. The purpose of this study was to evaluate kidney
carcinogenicity.
Cobalt chloride was tested in only one study by subcutaneous injection in male rats (Shabaan et
al. 1977) in two similar experiments, one that ended after 8 months and one that lasted for 12
months. Only the 12-month study included an untreated control, but it seems reasonable to use
that control for the 8-month study, especially since no neoplasms developed in the controls at 12
months. In the 12-month experiment, fibrosarcomas were found in 8/11 survivors at both the
subcutaneous injection sites (4) and at sites distant from the injection site (4). In the 8-month
experiment, 6 of the 16 animals who were alive at the end of the observation period had tumors
(Shabaan et al. 1977). (Animals who died before 8 or 12 months were not examined for tumors.)
Due to poor reporting, it was not possible to differentiate between tumors that occurred at
injection sites versus non-injection sites. The cobalt-exposed animals developed persistent
hyperlipaemia, and mortality was high for the treated animals.
Cobalt oxide was injected (i.p., s.c., i.m.) into rats in three studies (Steinhoff and Mohr 1991,
Gilman and Ruckerbauer 1962) and into mice (i.m.) in one study (Gilman and Ruckerbauer
1962). All rat studies reported significant increases in local neoplasms, either sarcoma,
histiocytoma, or both combined. Although few rats were used in the studies, more than 50% of
the rats developed injection site tumors. No treatment-related increase in neoplasms was found in
the one study in mice. The number of animals was adequate in this study; however, only one
dose was used (lower than the rat study) and there was little information on dose selection. There
were some concerns about potential for confounding from the animal husbandry conditions and
limited information on chemical purity in the studies in rats and mice by Gilman and
Ruckerbauer (1962). However, no tumors were observed in mice, the controls, or rats and mice
injected with thorium dioxide, thus arguing against any potential confounding.
Only one study tested cobalt sulfide, which was injected intrarenally into female rats (Jasmin and
Riopelle 1976). No neoplasms were reported in this study; however, the doses used in this study
may have been low since they were similar to the doses used in the study with cobalt metal that
was also negative.
82
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Table 5-4. Injection site neoplasms and non-neoplastic lesions in experimental animals exposed to cobalt compounds
Reference & year,
animal, study duration
Hansen 2006
Rat (Sprague-Dawley)
Male
12 mo
Substance & purity
Dosing regimen
Cobalt metal [nano and
bulk]
(Bulk metal particles
were 50-200 nm in size,
with an average of 120
nm. The surface area to
mass ratio was 4.75 and
for the nano-particles it
was 50,000, but the size
of the nano-particles
were not reported.)
Nano:
IM implant (left side of
vertebra)
Single dose
Bulk:
SC implant (right side
of vertebra)
Single dose
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Fibroblastic proliferation 6 months
0 cm²
4
0
Nano 2 cm²
4
2
Bulk 2 cm²
4
0
Sarcoma 6 months
0 cm²
4
0
Nano 2 cm²
4
1
Bulk 2 cm²
4
0
Fibroblastic proliferation 8 months
0 cm² (12 mo)
6
0/6 (0%)
Nano 2 cm²
6
1/6 (16.7%)
Bulk 2 cm²
6
1/6 (16.7%)
Sarcoma 8 months
Heath 1956
Rat (Hooded)
Male
life span
Cobalt metal
("Spectroscopically
pure," Particle size: 3.5
× 3.5 μm to 17 × 12
μm)
IM inj. in fowl serum
Single dose
0 cm² (12 mo)
6
0/6 (0%)
Nano 2 cm²
6
5/6 (83.3%)[**]
Bulk 2 cm²
6
0/6 (0%)
Rhabdomyofibrosarcoma or sarcoma combined
0 mg/rat
NR
0/10 (0%)
28 mg/rat
8
4/10 (40%)
This draft document should not be construed to represent final NTP determination or policy
Comments, strengths,
and limitations
4 animals (PVC control
and treated) sacrificed
at 6 months and the
remaining 6 animals
sacrificed at either 8
(treated) or 12 months
(PVC controls). Treated
animals sacrificed at 8
months due to
mortality.
Strengths: Tested
multiple materials in
addition to cobalt and
thus able to provide
information on whether
effects were due to
physical state.
Limitations: Inert
polyvinyl chloride
particles were used as a
negative control. Only a
small number of males
were tested at a single
dose level. Short
duration and unable to
fully evaluate effects
from cobalt bulk
particles.
No data was given on
the survival of untreated
controls. 2/10 treated
males without tumors
died before final
sacrificed.
83
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Strengths: Observation
duration was sufficient
and both sexes were
tested.
Limitations:
Incomplete reporting of
many elements. Limited
sensitivity due to only
one dose level and few
rats tested. Full
necropsies were not
reported.
Heath 1956
Rat (Hooded)
Female
life span
84
Cobalt metal
("Spectroscopically
pure," Particle size: 3.5
× 3.5 μm to 17 × 12
μm)
Series I and Series II
IM inj. in fowl serum
Single dose
Sarcoma (Rhabdomyofibrosarcoma or
fibrosarcoma)
0 mg/rat
NR
0/10 (0%)
28 mg/rat
Series I
6
5/10 (50%)
28 mg/rat
Series II
10
7/10 (70%)
This draft document should not be construed to represent final NTP determination or policy
No data were reported
on the survival of
untreated controls. For
treated animals, 4/10
rats (Series I) and 0/10
(Series II) without
tumors died before final
sacrificed.
Strengths: Observation
duration was sufficient
and both sexes were
tested.
Limitations:
Incomplete reporting of
many elements. Limited
sensitivity due to only
one dose level and few
rats tested. Full
necropsies were not
reported.
Other comments:
Series I used a
concurrent control, but
Series II used the same
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
controls, which was
non-concurrent. 6/7
sarcoma in Series I and
2/5 in Series II were
rhabdomyofibrosarcoma
Heath and Daniel
1962
Rat (Hooded)
Female
28 months
Cobalt metal
(Purity not reported,
Particle size: 3.5x3.5
μm to 17x12 μm)
Intrathoracic inj. (in
serum)
Single injection
Jasmin and Riopelle
1976
Rat (Sprague-Dawley)
Cobalt metal
NR
Intrarenal placement
Single dose
Mixed sarcoma intrathoracic region
0 mg/dosea
NR
0/10 (0%)
28 mg/dose
11
4/12 (33%)
Kidney neoplasm NOS
0 mg/rat
NR
0/16 (0%)
10 mg/rat
NR
0/18 (0%)
This draft document should not be construed to represent final NTP determination or policy
Survival was only
reported for exposed
rats, which was 12/20
on day 3 and 11/20 after
11 months.
Strengths: Observation
duration was sufficient
and both sexes were
tested.
Limitations: Historical
controls from Heath
1956 used because there
was no concurrent
control. Few animals
were used, and full
necropsies were not
done, only skin tumors
were histologically
examined. Incomplete
reporting of many
elements.
Other comments: 3 of
4 tumors originated in
part from cardiac
muscle, which are very
rare.
Survival was not
reported.
Strengths: Moderate
85
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Female
12 months
Shabaan 1977
Rat (Wistar)
Male
8 and 12 mo
86
Comments, strengths,
and limitations
number of animals.
Limitations: Only a
single dose level, which
was lower than other
studies, was tested in
only females.
Incomplete reporting
for many elements. Full
necropsies were not
performed, though the
abdominal and thoracic
cavities were examined.
Cobalt chloride
NR
SC inj.
1 dose/day × 5 days,
then 9 days off, then 1
dose/day × 5 days (total
19 days)
Injection site and non-injection fibrosarcoma
0 mg/kg bw 12
mo
19
0/19 (0%)
40 mg/kg bw 8
mo
16
6/16 (30%)[**]
40 mg/kg bw
12 mo
11
8/11 (40%)[***]
This draft document should not be construed to represent final NTP determination or policy
Treatment-related
decrease in survival;
16/20 survived at 8
months and 11/20
survived at 12 months.
Limitations: Exposure
resulted in persisent
hyperlipaemia and high
mortality. Animals
dying before the end of
observation period were
not exmained for
tumors. The tumors at
injection sites and noninjection sites weren’t
clearly reported
Other comments: No
concurrent untreated
controls used at 8
months, 12 months
controls used as
comparison group.
Statistical testing
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
(Fisher’s Exact Test)
reported by IARC.
Steinhoff and Mohr
1991
Rat (Sprague-Dawley)
Male and Female
life span
Cobalt oxide
("Chemically pure,"
80% of particles were
5–40 μm)
IP inj.
1 dose/2 mo × 6 mo
Sarcoma
0 mg/kg
NR
1/20 (5%)
200 mg/kg
NR
3/20 (15%)[*]
Mesothelioma
0 mg/kg
NR
0/20 (0%)
200 mg/kg
NR
1/20 (5%)
Histiocytoma
Steinhoff and Mohr
1991
Rat (Sprague-Dawley)
Male
life span
Cobalt oxide
("Chemically pure",
80% of particles were
5–40 μm)
SC inj.
1 inj/day, 5 day/week ×
730 days
0 mg/kg
NR
1/20 (5%)
200 mg/kg
NR
10/20 (50%)[**]
Histiocytoma or sarcoma combined
0 mg/kg/wk
NR
0/10 (0%)
0 mg/kg/wk
NR
0/10 (0%)
2 mg/kg ×
5/wk
NR
5/10 (50%)[*]
10 mg/kg/wk
NR
4/10 (40%)[*]
This draft document should not be construed to represent final NTP determination or policy
Survival was not
reported.
Strengths: Both sexes
of rats were tested with
a long duration of
observation.
Limitations:
Incomplete reporting.
Limited sensitivity
because of few animals
per group, only one
dose level was tested,
and exposure was for
less than one year.
Limited histological
examination
Other comments:
Results were reported as
combined for males and
females.
Survival was not
reported.
Strengths: Duration of
exposure and
observation were
sufficient. One dose
level was tested, at two
intensity levels and two
untreated control groups
used.
Limitations: Limited
sensitivity due to few
animals per group and
87
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
only males tested.
Limited histological
examination.
Incomplete reporting of
many elements.
Gilman and
Ruckerbauer 1962
Rat (Wistar)
Male and female
489 days
Cobalt oxide
(purity not reported,
particle size was < 5
μm)
IM inj.
Single dose
Gilman and
Ruckerbauer 1962
Mouse (Swiss)
Female
751 days
Cobalt oxide
(purity not reported,
particle size was < 5
μm)
IM inj.
Single dose
88
Sarcoma
0 mg/rat
10
0/10 (0%)
30 mg/rat
10
5/10 (50%)[*]
Sarcoma
0 mg/mouse
48
0/51 (0%)
20 mg/mouse
46
0/50 (0%)
This draft document should not be construed to represent final NTP determination or policy
Survival was similar to
control at 90 days.
Strengths: The
duration of observation
was sufficient and both
sexes were tested.
Limitations: Limited
sensitivity because only
a single dose was given
at one dose level and
few animals per group
were tested. Incomplete
reporting for many
elements. Animal
bedding was
periodically dusted with
rotenone powder.
Other comments:
Results were reported as
combined for males and
females.
Survival was similar to
control at 90 days.
Strengths: The
duration of observation
and the numbers of
animals per group were
sufficient.
Limitations: Limited
sensitivity due to only a
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
single dose was given at
one dose level, without
a rationale, to females
only. Half of the rats
were survivors from a
preliminary study who
received unwashed
cobalt, which was
known to contain other
toxic chemicals.
Bedding was
periodically dusted with
rotenone powder.
Incomplete reporting
for many elements.
Jasmin and Riopelle
1976
Rat (Sprague-Dawley)
Female
12 months
Cobalt sulfide
NR
Intrarenal placement
Single dose
Kidney neoplasm NOS
0 mg/rat
NR
0/16 (0%)
10 mg/rat
NR
0/20 (0%)
Survival was not
reported.
Strengths: Moderate
number of rats per
groups.
Limitations: Limited
sensitivity due to only a
single dose level, which
was lower than other
studies and only
females tested.
Incomplete reporting.
Full necropsies were not
performed, though the
abdominal and thoracic
cavities were examined.
* = P-value ≤ 0.05; ** = P-value ≤ 0.01; *** = P-value ≤ 0.001.
NR = Not reported.
+
= Number of animals at the beginning of the study, except for Hansen 2006 and Heath and Daniel 1962, which used the number of animals that were examined
at the time of sacrifice (Hansen 2006) or the number of animals that survived beyond day 4 (Heath and Daniel 1962).
[ ] = Statistical significance calculated by NTP using Fisher’s Exact Test.
This draft document should not be construed to represent final NTP determination or policy
89
06/05/15
a
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Historical control group from earlier study by the same author.
90
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Other and distal site neoplasms and non-neoplastic lesions
Several lines of evidence support systemic exposure of rats and mice to cobalt. Cobalt
concentrations and burdens increased with increasing exposure concentrations in all studies in all
tissues examined; however, tissue burdens normalized by exposure concentration showed
increased levels only in the liver (NTP 2014b; see Section 5.1.3). In addition, neoplasms were
observed at several organ sites (pancreas, hematopoietic system, and kidney distal to the route of
administration.
Adrenal gland
Neoplasms of the adrenal gland were reported in two inhalation studies testing cobalt metal and
cobalt sulfate (see Table 5-5) (NTP 2014b, 1998, Wehner et al. 1977). In the four NTP studies,
cobalt metal and cobalt sulfate heptahydrate were each tested in both mice and rats, but adrenal
gland neoplasms developed only in rats. One study reported a single adrenal gland neoplasm in
hamsters exposed to cobalt oxide (Wehner et al. 1977). There is a high background of adrenal
tumors in the male rats in the two NTP studies. Adrenal gland neoplasms can develop from
damage to lungs that cause obstructive sequela by causing systemic hypoxemia, leading to
chronic stimulation of catecholamine release by the adrenal medulla and subsequent neoplastic
development (NTP 2014b). Since inhalation of cobalt caused lesions in the lung that could cause
obstruction (chronic inflammation), it is possible that the adrenal glands are not directly caused
by systemic exposure to cobalt, but could be a secondary response to lung damage. However,
there is not enough evidence to differentiate between a direct or indirect cause of adrenal gland
neoplasms from cobalt exposure.
The strongest evidence for a treatment-related effect comes from the rat studies with cobalt
metal. Inhalation exposure to cobalt metal significantly increased bilateral malignant
pheochromocytoma in the high-dose group (5 mg/m3) and all malignant pheochromocytoma,
malignant or benign pheochromocytoma combined, and benign pheochromocytoma in both the
mid- (2.5 mg/m3) and high-dose groups in male rats. In females, there was a significantly
increased incidence of bilateral malignant pheochromocytoma as well as malignant
pheochromocytoma overall at the high dose and malignant or benign pheochromocytoma
combined, and bilateral benign pheochromocytoma as well as benign pheochromocytoma in both
the mid- and high-dose groups (NTP 2014b). Hyperplasia of the adrenal gland was also
significantly increased in females at mid and high doses, and was significantly decreased in
males in the mid- and high-dose groups.
Cobalt sulfate heptahydrate caused significant increases in malignant, benign, or complex
adrenal neoplasms combined in both sexes, which were higher than historical controls (NTP
1998). However, increases were only significant in the high-dose (3 mg/m3) group in females
and the mid-dose (1 mg/m3) group in males. Females had a significant trend of increasing tumor
incidence with increasing dose for benign pheochromocytoma and all tumor types combined.
Hyperplasia was significantly increased in females and the high-dose, but was significantly
decreased at the low-dose (0.3 mg/m3) males.
Wehner reported finding a single adrenal gland adenoma in the cortex of hamsters after
inhalation of cobalt oxide. (Wehner et al. 1977). Wehner only tested one dose level, 10 mg/m3,
which was higher than those used in mice or rats in the two NTP studies. The significant
This draft document should not be construed to represent final NTP determination or policy
91
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
increases in rats, but not mice or hamsters, could indicate a species difference in sensitivity to
developing adrenal gland tumors from cobalt exposure, especially considering hamsters received
a higher dose level than the rats.
Distal sites: Pancreatic islet cell, hematopoietic system, and kidney
Inhalation exposure to cobalt metal also caused other tumors at sites distant from the route of
administration: pancreas in male rats and mononuclear-cell leukemia in female rats in the NTP
inhalation bioassay of cobalt metal (Behl et al. 2015, NTP 2014b). A non-significant increase in
the incidence of kidney tumors was observed in male rats. It is not clear whether the kidney
tumors were treatment related. Tumors were not observed in the pancreas, kidney, or
hematopoietic system of rats exposed to cobalt sulfate or mice exposed to either form of cobalt.
Findings are presented in Table 5-5 and briefly summarized below.
Male rats exposed to cobalt metal were found to have a significant increase in the incidences of
pancreatic islet cell carcinoma or adenoma combined in both the mid- and high-dose groups and
a significant positive dose-related trend was observed. A significant increase in the incidence of
pancreatic adenoma was also observed in the mid-dose group in males. The non-significant
increases in the incidence of pancreatic islet cell carcinomas observed in female rats exceeded
the historical controls for all routes of administration and thus may have been related to
exposure. However, historical controls were limited as they were based on a dataset of only 100
Fischer 344/NTac rats from two NTP carcinogenicity studies. Significant increases in the
incidence of mononuclear-cell leukemia were seen in females in all dose groups, which exceeded
the limited historical controls for all exposure routes. The onset of leukemia in females was
shorter in cobalt-exposed groups, but no statistical calculations were done to tell if the
differences were significant. The incidence of mononuclear-cell leukemia was similar in male
rats compared to the untreated controls.
The incidence of kidney neoplasms (adenoma or carcinoma combined) was higher (although not
significantly so) in the low- and high-dose male rats compared to the concurrent controls and a
significant trend was observed. The incidence exceeded the historical controls for all routes of
administration, but the historical controls are limited as mentioned above. Four of the five
neoplasms were adenomas. In analyses of standard and extended evaluations, a significant trend
was observed; two of the seven neoplasms in the high-dose group were carcinomas. Kidney
neoplasms are relatively rare, so non-significant increases may be related to cobalt exposure
(NTP 2014b). No treatment-related non-neoplastic lesions were observed. Two studies injected
cobalt sulfide or cobalt metal directly into the kidneys of female rats in one publication (Jasmin
and Riopelle 1976). No kidney tumors or any other tumors were reported as being significantly
increased. Only a single dose was given at one dose level and the dose was lower than that used
in other injection studies.
92
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Table 5-5. Other and distal site neoplasms and relevant non-neoplastic lesions in experimental animals exposed to cobalt compounds
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Adrenal gland
NTP 2014b
Rat (F344/NTac)
Male
105 wk
Bilateral malignant pheochromocytoma
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
0/50 (0%)
2.5 mg/m³
16
0/50 (0%)
5 mg/m³
16
7/50 (14%)**
Malignant pheochromocytomaa
0 mg/m³
17
2/50 (5%)
1.25 mg/m³
20
2/50 (5%)
2.5 mg/m³
16
9/50 (21%)*
5 mg/m³
16
16/50 (39%)***
Trend-test P-value: 0.001
Benign pheochromocytomaa
0 mg/m³
17
15/50 (36%)
1.25 mg/m³
20
23/50 (54%)
2.5 mg/m³
16
37/50 (81%)***
5 mg/m³
16
34/50 (76%)***
Survival was similar to
controls.
Strengths: A welldesigned study in all
factors.
Limitations: Decreases
in body weight in mid
and high dose rats.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
carcinogenicity studies
and so it is based on
only 100 rats.
Trend-test P-value: 0.001
Malignant or benign combined
pheochromocytomaa
0 mg/m³
17
17/50 (40%)
1.25 mg/m³
20
23/50 (54%)
2.5 mg/m³
16
38/50 (83%)***
5 mg/m³
16
41/50 (91%)***
Trend-test P-value: 0.001
This draft document should not be construed to represent final NTP determination or policy
93
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
NTP 2014b
Rat (F344/NTac)
Female
105 wk
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Bilateral malignant pheochromocytoma
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
1/50 (2%)
2.5 mg/m³
24
1/49 (2%)
5 mg/m³
25
4/50 (8%)*
Malignant pheochromocytomaa
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
2/50 (5%)
2.5 mg/m³
24
3/49 (8%)
5 mg/m³
25
11/50 (27%)***
Trend-test P-value: 0.001
Bilateral benign pheochromocytoma
0 mg/m³
35
2/50 (4%)
1.25 mg/m³
26
4/50 (8%)
2.5 mg/m³
24
8/49 (16%)*
25
19/50 (38%)**
5 mg/m³
Benign pheochromocytomaa
0 mg/m³
35
6/50 (14%)
1.25 mg/m³
26
12/50 (27%)
2.5 mg/m³
24
22/49 (52%)***
5 mg/m³
25
36/50 (81%)***
Trend-test P-value: 0.001
Malignant or benign combined
pheochromocytomaa
0 mg/m³
94
35
6/50 (14%)
1.25 mg/m³
26
13/50 (29%)
2.5 mg/m³
24
23/49 (55%)***
5 mg/m³
25
40/50 (89%)***
This draft document should not be construed to represent final NTP determination or policy
Comments, strengths,
and limitations
Survival was
significantly decreased
in the mid-dose group.
Strengths: A welldesigned study in
almost all factors.
Limitations: A
significant decrease in
survival of female rats.
Decreases in body
weight in mid- and
high-dose rats.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
carcinogenicity studies
and so it is based on
only 100 rats.
Significantly increased
non-neoplastic lesions:
Hyperplasia - low and
medium
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Trend-test P-value: 0.001
NTP 1998
Rat (F344)
Male
2 yr
Cobalt sulfate
(99% pure)
Inhalation
6 hr/day, 5 days/wk ×
105 wk
Benign pheochromocytomab
0 mg/m³
17
14/50 (51%)
0.3 mg/m³
15
19/50 (70%)
1.0 mg/m³
21
23/48 (72%)
3.0 mg/m³
15
20/50 (71%)
Survival was similar to
controls.
Strengths: A welldesigned study in all
factors and survival was
similar to controls..
Limitations: None.
Malignant, benign, or complex
pheochromocytoma combinedb
NTP 1998
Rat (F344)
Female
2 yr
Cobalt sulfate
(99% pure)
Inhalation
6 hr/day, 5 days/wk ×
105 wk
0 mg/m³
17
15/50 (52%)
0.3 mg/m³
15
19/50 (70%)
1.0 mg/m³
21
25/48 (74%)*c
3.0 mg/m³
15
20/50 (71%)
Benign pheochromocytomab
0 mg/m³
28
2/48 (5%)
0.3 mg/m³
25
1/49 (3%)
1.0 mg/m³
26
3/50 (9%)
3.0 mg/m³
30
8/48 (26%)*
Trend-test P-value: 0.004
Malignant, benign, or complex combinedb
0 mg/m³
28
2/48 (4%)
0.3 mg/m³
25
1/49 (2%)
1.0 mg/m³
26
4/50 (8%)
3.0 mg/m³
30
10/48 (21%)*d
Survival was similar to
controls.
Strengths: A welldesigned study in all
factors and survival was
similar to controls.
Limitations: None.
Other comments:
Significantly increased
non-neoplastic lesions:
Adrenal gland:
Hyperplasia - high dose
Trend-test P-value: 0.001
This draft document should not be construed to represent final NTP determination or policy
95
06/05/15
Reference & year,
animal, study duration
Wehner 1977
Hamster (Syrian
Golden, random bred
ENG:ELA)
Male
Lifespan
96
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Cobalt oxide
(Purity not reported, the
median diameter of
particles was 0.14 μm,
with a median mass
diameter of 0.45 μm
and a geometric
standard deviation of
1.9 μm)
Inhalation
7 hr/day, 5 days/wk ×
lifespan
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Adenoma (cortex)
0 µg/L
NR
0/51 (0%)
10.1 µg/L
NR
1/50 (2%)
This draft document should not be construed to represent final NTP determination or policy
Comments, strengths,
and limitations
Survival in exposed
group is similar to
control but is poor in
both groups.
Strengths: Duration of
exposure and
observation were
sufficient.
Limitations:
Incomplete reporting.
Low sensitivity because
of relatively poor
survival of both
exposed and controls,
only a single dose level
was tested with no
justification for
choosing that dose
level.
Other comments: The
study looked at cobalt’s
effect on cigarette
smoke, but a cobalt
oxide only group was
tested. Cobalt-exposed
hamsters developed
pneumoconiosis.
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Pancreas
NTP 2014b
Rat (F344/NTac)
Male
105 wk
Carcinomaa
0 mg/m³
17
2/50 (5%)
1.25 mg/m³
20
1/50 (3%)
2.5 mg/m³
16
5/48 (13%)e
5 mg/m³
16
6/49 (15%)e
Trend-test P-value: 0.021
Adenomaa
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
1/50 (3%)
2.5 mg/m³
16
6/48 (15%)*
5 mg/m³
16
3/49 (8%)
Carcinoma or adenoma combineda
0 mg/m³
17
2/50 (5%)
1.25 mg/m³
20
2/50 (5%)
2.5 mg/m³
16
10/48 (25%)*e
5 mg/m³
16
9/49 (23%)*e
Survival was similar to
controls.
Strengths: A welldesigned study in all
factors.
Limitations:
Decreases in body
weight in mid- and
high-dose rats.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
carcinogenicity studies
and so it is based on
only 100 rats.
Trend-test P-value: 0.002
This draft document should not be construed to represent final NTP determination or policy
97
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
NTP 2014b
Rat (F344/NTac)
Female
105 wk
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Carcinomaa
0 mg/m³
35
1/50 (2%)
1.25 mg/m³
26
0/50 (0%)
2.5 mg/m³
24
0/50 (0%)
5 mg/m³
25
3/50 (7%)f
Adenoma
0 mg/m³
35
0/50 (0%)
1.25 mg/m³
26
0/50 (0%)
2.5 mg/m³
24
0/50 (0%)
5 mg/m³
25
1/50 (2%)
Carcinoma or adenoma combineda
98
0 mg/m³
35
1/50 (2%)
1.25 mg/m³
26
0/50 (0%)
2.5 mg/m³
24
0/50 (0%)
5 mg/m³
25
3/50 (7%)f
This draft document should not be construed to represent final NTP determination or policy
Comments, strengths,
and limitations
Survival was
significantly decreased
in the mid-dose group.
Strengths: A welldesigned study in
almost all factors.
Limitations: A
significant decrease in
survival of female rats.
Decreases in body
weight in mid- and
high-dose rats.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
NTP carcinogenicity
studies and so it is
based on only 100 rats.
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference & year,
animal, study duration
Substance & purity
Dosing regimen
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
06/05/15
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Hematopoietic system
NTP 2014b
Rat (F344/NTac)
Male
105 wk
NTP 2014b
Rat (F344/NTac)
Female
105 wk
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Mononuclear cell leukemiaa
0 mg/m³
17
21/50 (49%)
1.25 mg/m³
20
25/50 (58%)
2.5 mg/m³
16
22/50 (50%)
5 mg/m³
16
22/50 (48%)
Mononuclear cell leukemiaa
0 mg/m³
35
16/50 (36%)
1.25 mg/m³
26
29/50 (62%)**g
2.5 mg/m³
24
28/50 (61%)*g
5 mg/m³
25
27/50 (59%)*g
This draft document should not be construed to represent final NTP determination or policy
Survival was similar to
controls.
Strengths: A welldesigned study in all
factors.
Limitations: None.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
carcinogenicity studies
and so it is based on
only 100 rats.
Survival was
significantly decreased
in the mid-dose group.
Strengths: A welldesigned study in
almost all factors.
Limitations: A
significant decrease in
survival of female rats.
Decreases in body
weight in mid- and
high-dose rats.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
NTP carcinogenicity
studies and so it is
based on only 100 rats.
99
06/05/15
Reference & year,
animal, study duration
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance & purity
Dosing regimen
Cobalt metal
(98% pure, mass
median aerodynamic
diameter 1–3 μm)
Inhalation
6 hr/day, 5 day/wk ×
105 wk
Dose levels
# animals
at sacrifice
Tumor incidence
(n/N+) (%)
Comments, strengths,
and limitations
Kidney
NTP 2014b
Rat (F344/NTac)
Male
105 wk
Tubule adenomaa
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
1/50 (3%)h
2.5 mg/m³
16
0/50 (0%)
5 mg/m³
16
3/50 (8%)h
Tubule carcinoma or adenomaa
0 mg/m³
17
0/50 (0%)
1.25 mg/m³
20
1/50 (3%)h
2.5 mg/m³
16
0/50 (0%)
5 mg/m³
16
4/50 (10%)h
Trend-test P-value: 0.018
Tubule carcinoma or adenomaai
0 mg/m³
17
3/50 (8%)
1.25 mg/m³
20
1/50 (3%)
2.5 mg/m³
16
1/50 (2%)
5 mg/m³
16
7/50 (17%)
Trend-test P-value: 0.023
*P-value < 0.05; **P-value < 0.01; ***P-value < 0.01.
+
= Number of animals necropsied for NTP 2014b and NTP 1998 and is the number of animals at the beginning of the study for all other studies.
NR = Not reported.
a
Adjusted percent incidence based on Poly-3 estimated neoplasm incidence after adjustment for intercurrent mortality.
b
Adjusted percent incidence based on Kaplen-Meier estimated incidence at the end of the study after adjustment for intercurrent mortality.
c
Increased over historical control levels with a mean of 176/623 and range of 8% to 50%.
d
Increased over historical control levels with a mean of 39/608 and range of 2% to 14%.
e
Increased over historical control levels with a mean of 2/100 and range of 0% to 4%.
f
Increased over historical control levels with a mean of 1/100 and range of 0% to 2%.
g
Increased over historical control levels with a mean of 35/100 and range of 32% to 38%.
h
Increased over historical control levels with a mean of 1/100 and range of 0% to 2%.
i
Analyzed by standard and extended evaluation.
100
This draft document should not be construed to represent final NTP determination or policy
Survival was similar to
controls.
Strengths: A welldesigned study in all
factors.
Limitations: Decreases
in body weight in midand high-dose rats.
Other comments:
Historical controls were
limited, as Fischer
344/NTac rats have
only been used in two
NTP carcinogenicity
studies and so it is
based on only 100 rats.
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
5.2
5.2.1
06/05/15
Co-carcinogenicity studies
Overview of the studies
Nine co-carcinogen studies were identified that tested soluble compounds, including four studies
using cobalt chloride (Zeller 1975, Finogenova 1973, O'Hara et al. 1971, Kasirsky et al. 1965)
and three studies using sodium cobaltinitrite (O'Hara et al. 1971, Thompson et al. 1965,
Orzechowski et al. 1964); and a poorly soluble compound, cobalt oxide, in two studies
(Steinhoff and Mohr 1991, Wehner et al. 1977) (see Table 5-6). Most co-carcinogen studies were
conducted in mice, though two studies were conducted in rats (Steinhoff and Mohr 1991, Zeller
1975) and one study conducted in hamsters (Wehner et al. 1977). Almost all of the cocarcinogen studies used dermal exposure to methylcholanthrene as the known carcinogen, with
Zeller using subcutaneous injections of diethylnitrosamine, Steinhoff and Mohr using
intratracheal instillation of benzo[a]pyrene, and Wehner using inhalation exposure to cigarette
smoke. Methylcholanthrene induced skin tumors, while diethylnitrosamine induced liver and
nasal tumors, benzo[a]pyrene induced lung tumors, and cigarette smoke increased incidences of
total malignant or total benign neoplasms. Cobalt compounds were administered by
intraperitoneal injection in all but four studies, which used subcutaneous injection (Zeller 1975),
drinking water (Thompson et al. 1965), inhalation (Wehner et al. 1977), and intratracheal
instillation (Steinhoff and Mohr 1991) as routes of exposure.
Table 5-6. Overview of co-carcinogenicity studies in experimental animals reviewed
Co—carcinogen &
route
Exposure period/
study duration
diethylnitrosamine
SC inj.
methylcholanthrene
dermal
43 wk/
lifespan
8 wk/
8wk
(Zeller 1975)
IP inj.
methylcholanthrene
dermal
5 wk/
17 wk
(O'Hara et al.
1971)
Cobalt
chloride
IP inj.
methylcholanthrene
dermal
10 wk/
10wk
(Kasirsky et al.
1965)
Mouse CF-1
(M&F)
Sodium
cobaltinitrite
IP inj.
methylcholanthrene
dermal
5 wk/
17 wk
(O'Hara et al.
1971)
Mouse CF-1
(M&F)
Sodium
cobaltinitrite
Drinking
water
methylcholanthrene
dermal
11wk/1
1wk
(Thompson et
al. 1965)
Mouse CF-1
(M&F)
Sodium
cobaltinitrite
IP inj.
methylcholanthrene
dermal
72 days/
75 days
(Orzechowski et
al. 1964)
Rat SpragueDawley (F)
Cobalt(II)
oxide
Intratracheal
instill.
benzo[a]pyrene
intratracheal instill.
47 wk/
lifespan
(Steinhoff and
Mohr 1991)
Hamster Syrian
Golden (M)
Cobalt(II)
oxide
Inhalation
cigarette smoke
inhalation
Lifespan/
lifespan
(Wehner et al.
1977)
Strain (sex)
Substance
Route
Rat Wistar
(M&F)
Mouse
CBAxC57B1 (F)
Cobalt
chloride
Cobalt
chloride
SC inj.
Mouse CF-1
(M&F)
Cobalt
chloride
Mouse CF-1
(M&F)
IP inj.
Reference
(Finogenova
1973)
M = male, F = female, instill. = instillation, inj. = injection, IP = intraperitoneal, IM = intramuscular, SC = subcutaneous, wk =
week, yr = year.
This draft document should not be construed to represent final NTP determination or policy
101
06/05/15
5.2.2
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Overview of the assessment of study quality and utility
Each of these primary studies was systematically evaluated for its ability to inform the cancer
hazard similar to that described for the carcinogenicity studies in Section 5.1.2. O’Hara et al.
(1971) conducted two co-carcinogenicity studies (one using cobalt chloride and the other using
sodium cobaltinitrite) that were considered inadequate to be used to evaluate the carcinogenicity
of cobalt, because they did not test the influence of cobalt on tumor formation, as cobalt was not
administered until after neoplasms were already detectable. No critical concerns were identified
in the remaining studies although they were considered to be of low quality. Finogenova (1973)
did not report neoplasm incidences, but did report neoplasm onset and latency. The other studies
had poor reporting of duration, survival, and results, as they were not reported for each gender,
but had combined data for both sexes. The study quality assessment is discussed in Appendix D.
All co-carcinogenicity studies were categorically restricted to being ranked no higher than “low”
for the utility to inform the carcinogenicity evaluation. This restriction was applied to account for
the indirect measure of carcinogenicity that co-carcinogenicity studies provide.
5.2.3
Assessment of findings from co-carcinogenicity studies
Co-carcinogenicity studies are also divided by site of neoplasm development into skin, lung,
liver, nasal neoplasms, and neoplasms of unspecified location. Only one co-carcinogen study
demonstrated an increased incidence of lung neoplasms from cobalt (cobalt oxide), while three
studies showed no effect from cobalt (cobalt chloride and cobalt oxide) and three studies
reported a decrease (cobalt chloride and sodium cobaltinitrite) in neoplastic incidence with the
additional exposure to cobalt compounds.
Skin
Four co-carcinogenicity studies of cobalt and methylcholanthrene were reviewed (Finogenova
1973, O'Hara et al. 1971, Kasirsky et al. 1965, Thompson et al. 1965, Orzechowski et al. 1964).
In all of the studies, methylcholanthrene was applied dermally to mice and either sodium
cobaltinitrite or cobalt chloride was administered in drinking water or by i.p injection. All studies
reported skin squamous-cell carcinoma (Finogenova was translated from Russian and was
reported as skin cancer NOS). Skin tumor incidences were reduced by co-administration of
cobalt in three of the four studies (Kasirsky et al. 1965, Thompson et al. 1965, Orzechowski et
al. 1964). In the fourth study, no differences were seen in the onset or latency of neoplasm
development for either skin “cancer NOS” or papilloma from the addition of cobalt chloride
(Finogenova 1973). The authors didn’t report any tumor incidences.
Lung
Two co-carcinogenicity studies used either inhalation or intratracheal instillation as the route of
exposure for both the cobalt compound and the known carcinogen (Steinhoff and Mohr 1991,
Wehner et al. 1977). Steinhoff and Mohr administered benzo[a]pyrene and cobalt oxide to
female rats by intratracheal instillation. The addition of cobalt oxide increased the incidence of
squamous-cell carcinoma of the lung (Steinhoff and Mohr 1991). An adenocarcinoma was also
reported in the group exposed to both compounds, but not in the group exposed to just
benzo[a]pyrene. However, the incidence of adenocarcinoma was not significantly increased by
cobalt oxide. Wehner exposed male hamsters to cigarette smoke and cobalt oxide by inhalation
102
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
(Wehner et al. 1977). No significant change in tumor incidence from the addition of cobalt oxide
was reported, but the locations of the neoplasms were not clearly reported.
Liver and nose
Only one co-carcinogen study reported neoplasms of the liver and nose (Zeller 1975). In this
study, the known carcinogen, diethylnitrosamine, was subcutaneously injected together with
cobalt chloride into male and female rats. Diethylnitrosamine induced neoplasms of the nose
(esthesioneuroepithelioma, poorly differentiated carcinoma NOS, and squamous-cell carcinoma)
and liver (hepatoma NOS, hepatocellular carcinoma, and cholangioma), but the addition of
cobalt chloride had no effect on the incidences.
Unspecified neoplasm or non-neoplastic lesion locations
Only one co-carcinogen study reported neoplasms that were not specified as to their location or
even their histological type (Wehner et al. 1977). Significant decreases in the incidences of
neoplasms in cigarette smoke-exposed groups were seen with the addition of cobalt oxide.
Groups that were exposed to cobalt and cigarette smoke also had significantly lower body
weights than those exposed to just cigarette smoke, which might account for the lower neoplasm
incidence. This co-carcinogen study included a cobalt oxide alone group, which did not show a
significant increase in neoplasm incidence above that of untreated controls.
5.3
Synthesis of the findings across studies
Strengths of the available dataset include testing of cobalt compounds with different properties
such as particle versus salt and poorly soluble vs. readily soluble compounds. For some
compounds, several studies were available including robust studies with high utility for
evaluating carcinogenicity; importantly these include inhalation studies on both a water-soluble
(cobalt sulfate) and poorly soluble species (cobalt metal). For other cobalt compounds, there
were few studies, some of which were of more limited utility. The overall results for the
carcinogenicity studies are summarized by cobalt compound in Table 5-7.
In general the injection studies were less robust than the inhalation studies. Occupational
exposure to cobalt compounds usually occurs by inhalation and not by injection. However, the
injection route may be relevant to human exposure, in that cobalt is used in many types of
surgical implant materials. The interpretation of the carcinogenicity of the injection studies is
limited because many different types of particles or metals including substances that are
considered to be relatively inert have induced tumors in rats (IARC 2006). Nevertheless, in two
studies, no tumors were observed after implantation or injection of other materials (titanium
dioxide, silicon dioxide, or thorium dioxide), whereas tumors were observed after implantation
of cobalt metal (Hansen et al. 2006) or injection of a cobalt compound, cobalt oxide (Gilman and
Ruckerbauer 1962). Hansen et al. found that the materials (titanium dioxide and silicon dioxide)
that did not induce neoplasms had the same physical characteristics (i.e., surface to volume ratio)
as those that did (cobalt and nickel), which suggests that the tumors were due to carcinogenic
properties of cobalt and not just to a reaction to any physical implant. Overall, the injection
studies are considered to provide supporting evidence for the carcinogenicity of cobalt.
Most of the neoplasms induced by cobalt compounds occur at the site of administration. Lung
tumors are only seen in inhalation or intratracheal instillation studies and tissue sarcoma
This draft document should not be construed to represent final NTP determination or policy
103
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
developed in the local tissue at the sites of injection. Both the lung tumors from inhalation and
tissue sarcomas from injections were caused by different cobalt forms including cobalt metal, a
poorly soluble compound (cobalt oxide) and two water-soluble compounds (cobalt sulfate for
lung tumors and cobalt chloride for injection tumors). In addition, cobalt metal induced several
types of tumors distal from the site of administration that were not caused by the other cobalt
species (with the possible exception of adrenal tumors from cobalt sulfate) although most of the
cobalt compounds were not adequately tested in models to evaluate these sites.
The most widely studied form of cobalt was cobalt metal. Lung tumors were observed in rats and
mice in both sexes after inhalation exposure (NTP 1998, 2014b), and injection site sarcomas
(primarily rhabdomyofibrosarcoma, fibrosarcoma or sarcoma) were observed in male and female
rats in several studies injecting cobalt metal by different methods (i.m. or intrathoracic) (Heath et
al. 1956, Heath and Daniel 1962). In addition, inhalation exposure to cobalt metal also increased
the incidences of adrenal gland tumors and tumors at distal sites – mononuclear-cell leukemia
and pancreas, and possibly kidney tumors (NTP 2014b). Cobalt metal nanoparticles, when
administered by i.m. injection, caused sarcoma in male rats; however, no inhalation studies were
identified (Hansen et al. 2006).
Similarly, a poorly soluble cobalt compound (cobalt oxide) caused both lung neoplasms (after
intratracheal instillation) in male rats and sarcoma and histiocytoma in several studies of male
and/or female rats after injection by various methods (s.c., i.m., i.p.) (Steinhoff and Mohr 1991,
Gilman and Ruckerbauer 1962). Inhalation exposure to cobalt oxide did not increase the
incidences of lung tumors in Syrian Golden Hamsters, but the hamster is a less sensitive model
for evaluating lung carcinogenicity (McInnes et al. 2013, Steinhoff and Mohr 1991) than the rat
or mouse. No tumors were observed in the only study of another poorly soluble cobalt
compound, cobalt sulfide, after intrarenal injection, but there were concerns about the dose level
in that study (Jasmin and Riopelle 1976).
Finally, consistent findings are also found for soluble cobalt salts. Inhalation exposure to cobalt
sulfate heptahydrate caused lung tumors in rats and mice and adrenal tumors in female rats.
Adrenal gland tumors were also induced by exposure to cobalt sulfate (NTP 1998). Although no
injection studies were identified that tested cobalt sulfate heptahydrate, a subcutaneous study of
cobalt chloride provided suggestive evidence that cobalt causes fibrosarcoma at the site of
administration and possibly at sites distant from the sites of administration; however, the
confidence in the evidence is reduced somewhat because of possible inadequate reporting or
procedures (Shabaan et al. 1977).
Co-carcinogenicity studies overall provided little if any support for the co-carcinogenicity of
cobalt compounds. One study reported that cobalt enhanced carcinogenicity, but the remaining
co-carcinogenicity studies reported either no effect or a decrease in carcinogenicity with coexposure to cobalt.
104
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Table 5-7. Overall results of carcinogenicity studies in experimental animals sorted by cobalt compound
Substance
Cobalt metal
Cobalt metal
Strain (sex)
Route
Exposure
period/
study duration
Rat
F344/NTac
(M&F)
Inhalation
2 yr/2 yr
Results
Reference
Lung
Alveolar/bronchiolar adenoma and
carcinoma M&F
(NTP
2014b)
Squamous-cell tumors (primarily
cystic keratinizing epithelioma) F;
[Equivocal] M
Mononuclear-cell leukemia F
Adrenal gland
Benign and malignant
pheochromocytoma M&F
Pancreas
Islet-cell adenoma or carcinoma
M;
[Equivocal: carcinoma] F
Kidney
Adenoma or carcinoma combined
[Equivocal] M
Cobalt metal
Mouse
B6C3F1/N
(M&F)
Inhalation
2 yr/2 yr
Lung
Alveolar/bronchiolar adenoma and
carcinoma M&F
(NTP
2014b)
Cobalt metal
[Nano]
Rat
SpragueDawley
(M)
IM inj.
Single dose/1
yr
Injection site
Sarcoma M
(Hansen et
al. 2006)
Cobalt metal
[Bulk]
Rat
SpragueDawley
(M)
SC inj.
Single dose/1
yr
Negative
Fibroblastic proliferation (nonneoplasia)
(Hansen et
al. 2006)
Cobalt metal
Rat
SpragueDawley (F)
Intrarenal
inj.
Single dose/ 1
yr
Negative
(Jasmin and
Riopelle
1976)
Cobalt metal
Rat
Hooded (F)
Intrathoraci
c
Single
dose/2.3 yr
Injection-site sarcoma [including
rhabdomyosarcoma of cardiac and
intercostal muscle, mixed
(Heath and
Daniel
1962)
Cobalt metal
Rat
Hooded
(M&F)
IM inj.
Single
dose/lifespan
Injection-site sarcoma
[rhabdomyofibrosarcoma M&F;
sarcoma M; fibrosarcoma F
(Heath
1956)
Inhalation
2 yr/2 yr
Lung
(NTP 1998)
Soluble cobalt compounds
Cobalt
Rat F344/N
This draft document should not be construed to represent final NTP determination or policy
105
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Substance
Strain (sex)
sulfate
heptahydrate
(M&F)
Route
Exposure
period/
study duration
Results
Reference
Alveolar/bronchiolar adenoma and
carcinoma M&F
Adrenal
Benign or malignant
pheochromocytoma F
Cobalt
sulfate
heptahydrate
Mouse
B6C3F1
(M&F)
Inhalation
2 yr/2 yr
Lung
Alveolar/bronchiolar adenoma and
carcinoma M&F
(NTP 1998)
Cobalt
chloride
Rat Wistar
(M)
SC inj.
8–12 mo/8–12
mo
Injection site
Fibrosarcoma M
(Shabaan et
al. 1977)
Non-injection site
Fibrosarcoma M
Poorly soluble cobalt compounds
Cobalt oxide Rat
Intratrachea
l instill.
SpragueDawley
(M&F)
1.5 yr/lifespan
Lung
Alveolar/bronchiolar carcinoma,
benign squamous epithelial
neoplasm, or alveolar/bronchiolar
adenoma combined M
(Steinhoff
and Mohr
1991)
Cobalt oxide
Rat
SpragueDawley
(M&F)
IP inj.
6 mo/lifespan
Injection site
Histiocytoma and sarcoma M&F
(Steinhoff
and Mohr
1991)
Cobalt oxide
Rat
SpragueDawley
(M)
SC inj.
730
day/lifespan
Injection site
Histiocytoma and sarcoma M
(Steinhoff
and Mohr
1991)
Cobalt oxide
Rat Wistar
(M&F)
IM inj.
Single
dose/1.3 yr
Injection site
Sarcoma M&F
Cobalt oxide
Mouse
Swiss (F)
IM inj.
Single dose/2
yr
Negative
Cobalt oxide
Hamster
Syrian
Golden (M)
Inhalation
Lifespan/lifesp
an
Negative
(Gilman and
Ruckerbaue
r 1962)
(Gilman and
Ruckerbaue
r 1962)
(Wehner et
al. 1977)
Cobalt
sulfide
Rat
SpragueDawley (F)
Intrarenal
inj.
Single dose/1
yr
Negative
106
(Jasmin and
Riopelle
1976)
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
6 Mechanistic and Other Relevant Effects
This section discusses the relative role of cobalt ions and particles in cobalt toxicity (Section
6.1), several proposed modes of actions for cobalt carcinogenicity (Section 6.2), cobalt levels in
neoplastic and non-neoplastic tissue from cancer patients (Section 6.3), and a synthesis (Section
6.4). Although the mechanism(s) of action for the reported cobalt-induced carcinogenic effects
are not completely understood, the experimental support for several possible modes of action,
including genotoxicity, is reviewed below. The genetic and related effects of cobalt and cobalt
compounds are reviewed in Appendix E.
6.1
Cobalt particles and cobalt ions
Studies with toxic metals in general show that solubility and particle size can play an important
role in metal-induced toxicity, genotoxicity, and carcinogenicity (Smith et al. 2014). The main
cobalt compounds studied for toxicological effects (including both micro- and nanoparticles) are
metallic cobalt (Co(0)), cobalt(II) oxide (CoO), cobalt(II,III) oxide (Co3O4), and various
cobalt(II) salts (e.g., cobalt sulfate, cobalt chloride) (Lison 2015, Ortega et al. 2014, Sabbioni et
al. 2014, Smith et al. 2014, Beyersmann and Hartwig 2008). Many cobalt(II) salts are readily
soluble in water and biological fluids (see Section 1).
Cobalt particles and ions induce similar biological effects in vivo and in vitro (e.g., cytotoxicity,
genotoxicity, apoptosis, and at high concentrations, necrosis with an inflammatory response)
(Smith et al. 2014, Simonsen et al. 2012). In particular, the effects of chronic exposure to cobalt
and cobalt compounds on the respiratory system in humans and experimental animals are well
documented (IARC 2006, ATSDR 2004, IARC 1991). Effects include respiratory irritation,
diminished pulmonary function, asthma, and interstitial fibrosis. Respiratory effects have been
observed in workers employed in cobalt refineries, hard metal workers, diamond polishers, and
ceramic dish painters. Several in vitro studies that specifically compared the cellular uptake
and/or molecular and cellular effects (e.g., cytotoxicity, genetic toxicity, ROS production) of
cobalt ions and particles (i.e., cobalt metal nanoparticles or cobalt oxide micro or nanoparticles
are shown in Table 6-1.
In vitro studies generally show that cobalt nanoparticles are more toxic than cobalt
microparticles due to increased surface reactivity resulting from a higher surface area/volume
ratio (Simonsen et al. 2012, Mo et al. 2008, Peters et al. 2007, Zhang et al. 2000). In addition,
relatively soluble cobalt particles (e.g., cobalt metal) are generally more cytotoxic and genotoxic
than cobalt ions (Sabbioni et al. 2014, Ponti et al. 2009, Peters et al. 2007) and cobalt ions are
generally more cytotoxic than cobalt particles with low solubility (e.g., cobalt oxide) (Table 6-1)
(Ortega et al. 2014, Smith et al. 2014, Papis et al. 2009). NTP (2009) previously reviewed
cobalt-tungsten carbide powders and hard-metals and reported that cobalt-tungsten carbide
particles were more cytotoxic and/or genotoxic than cobalt powder when tested in vivo (rat lung)
or in vitro in mammalian cells. The greater toxicity of cobalt-tungsten carbide was attributed to a
synergistic effect between the particles of cobalt and tungsten carbide that resulted in enhanced
production of ROS. Synergistic toxicity in vitro was also reported for cobalt with zinc (Bresson
et al. 2013) and cobalt with nickel (Patel et al. 2012) but not with chromium (Allen et al. 1997).
This draft document should not be construed to represent final NTP determination or policy
107
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 6-1. In vitro mechanistic data comparing effects of cobalt nanoparticles, microparticles, and ions
Reference
(Sabbioni et al.
2014)
Cobalt form (size, nm)
and cell type
Co NP (3.4)
Co MP (2,200)
CoCl2
IC50 µg/mL
Time Co MP Co NP
4h
12
19.5
12 h
10
10
24 h
11
10
48 h
10
9.9
Balb/3T3 mouse
fibroblasts
(Ortega et al.
2014)
Co3O4 MP (100-400)
CoCl2
BEAS-2B human lung
(Smith et al.
2014)
CoO MP (270–3,560)
CoCl2
WTHBF-6 human lung
fibroblasts
(Alarifi et al.
2013)
Co3O4 NP (21)
CoCl2
HepG2 human
hepatocarcinoma cells
108
Genotoxicitya
Cytotoxicity
ROS
Cellular uptake
Relative amount of Co
incorporated into the
DNA was
Co MP > Co NP > cobalt
ions.
No data
Co uptake was dose
dependent but
significantly higher for
NP and MP than for
cobalt ions. Maximum
uptake at 4 hours postexposure.
No data
No data
Co3O4 particles entered
cells via endocytosis and
released cobalt ions
within lysosomes over
long periods of time and
were responsible for
toxicity.
Both forms induced
concentration-dependent
increase in cytotoxicity;
however, similar levels of
cytotoxicity at intracellular
cobalt levels < 1,000 µM while
cobalt ions were more cytotoxic
than particulate Co at higher
levels.
Chromosome aberrations
(similar effect for
particulate and soluble
forms).
No data
Both particulate and
soluble Co induced a
concentration-dependent
increase in intracellular
cobalt ion levels.
Particle-cell contact was
required for uptake of
CoO.
Both forms induced
concentration-dependent
increase in cytotoxicity but
particulate Co was more
cytotoxic than soluble Co.
DNA damage (comet
assay, NP were more
potent than soluble form)
Particles induced ROS
and oxidative stress.
Effects were lower for
cobalt ions.
No data
IC25 µg/mL
IC50
IC75
Co3O4
50
170
600
Co2+
47
22
10
10
CoCl2
2.9
4.4
6.5
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference
(Horie et al.
2012)
Cobalt form (size, nm)
and cell type
CoO NP (> 10)
CoCl2
Cytotoxicity
Genotoxicitya
06/05/15
ROS
Cellular uptake
Both forms induced similar
concentration-dependent
increase in cytotoxicity in both
cell types.
No data
No increase in
intracellular ROS in cells
treated with cobalt ions
or particles.
No data
Both forms induced
concentration-dependent
increase in cytotoxicity but
cobalt ions were more toxic.
HepG2 cells not as sensitive
as ECV-304 cells.
No data
Particles but not ions
induced dose-dependent
increase in ROS
production in both cell
lines. HepG2 cells less
sensitive.
No data
No data
No data
Release of ROS was up
to 8 times higher for
particles than cobalt
ions.
No data
NPs induced a concentrationdependent reduction in all three
monocytic cell lines (prevented
by co-incubation with ascorbic
acid). CoCl2 at comparable
concentrations (50–350 µM)
was not cytotoxic.
No data
NPs induced ROS in a
concentration-dependent
manner in all cell lines
(prevented by both
ascorbic acid and
glutathione). CoCl2 did
not significantly increase
ROS.
No data
HaCaT human
keratinocytes
A549 human lung
carcinoma cells
(Papis et al.
2009)
Co3O4 NP (45)
CoCl2
HepG2 and ECV-304
human cell lines
(Limbach et al.
2007)
Co3O4 NP (20-75)
Co3O4/silica NP
Cobalt salt
A549 human lung
adenocarcinoma
epithelial cells
(Nyga et al.
2015)
CoNP (2-60)
CoCl2
U937 human monocytic
cell line, peripheral
blood mononuclear
cells, and alveolar
macrophages
This draft document should not be construed to represent final NTP determination or policy
109
06/05/15
Reference
(Annangi et al.
2014)
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Cobalt form (size, nm)
and cell type
CoNP (30.7 ± 20.2)
Ogg1+/+ and Ogg1-/mouse embryo
fibroblasts (MEF)
(Horev-Azaria et
al. 2011)
Co NP (10–50)
CoCl2
A549, NCIH441, Caco2, HepG2 (human lung,
colorectal, liver);
MDCK (dog kidney);
murine dendritic cells
(Ponti et al.
2009)
Co NP (20–500)
CoCl2
Balb/3T3 mouse
fibroblasts
(Kwon et al.
2009)
Co NP (30)
CoSO4
Genotoxicitya
ROS
Cellular uptake
NPs induced dose-dependent
cytotoxicity in wild-type and
knockout MEF cells (more
toxic to knockout cells).
Sub-toxic doses for 12
weeks induced cell
transformation (knockout
cells were more
sensitive).
Acute and subchronic
exposure induced ROS.
Greater toxicity in
knockout cells attributed
to increased sensitivity
to oxidative damage.
Dose-dependent increase
in cellular uptake of
CoNPs in wild-type and
knockout cells.
NPs and ions induced dosedependent cytotoxicity. NPs
were generally more toxic. Ion
sensitivity: A549 > MDCK >
NCIH441 > Caco-2 > HepG2 >
DC; NP sensitivity: A549 =
MDCK = NCIH441 = Caco-2 >
DC > HepG2. Toxicity of NP
aggregates attributed to
extracellular cobalt ion
dissolution (34%–44% at 48
and 72 hrs).
No data
No data
No data
Dose-dependent cytotoxicity
for both forms (higher for
particles at 2 and 24 h but
overlapping at 72 h).
Co NP induced DNA
damage, MN, and cell
transformation; CoCl2
induced DNA damage
only.
No data
No data
NPs and ions induced dosedependent cytotoxicity.
No data
No data
NP toxicity likely
resulted from cellular
uptake rather than
extracellular dissolution.
Cytotoxicity
RAW 264.7 murine
macrophages
110
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Reference
(Colognato et al.
2008)
Cobalt form (size, nm)
and cell type
Co NP (100–500)
CoCl2
Human peripheral
blood leukocytes
(Peters et al.
2007)
Co NP (28)
CoCl2
Human dermal
microvascular
endothelial cells
06/05/15
Cytotoxicity
Genotoxicitya
Co NP and cobalt ions induced
dose-related cytotoxic effects
(decrease in the cytokinesisblock proliferation index
(CBPI). CBPI was slightly
higher for ions at 10-5 M but
similar toxicity at >2 x 10-5 M
Cobalt ions induced clear
trend in increase of MN
frequency while Co NP
were less effective; MN
response varied with
donor. DNA damage with
NP only (comet assay,
short incubation time). No
MN observed at noncytotoxic concentrations.
No data
NP readily taken up by
cells. Cells exposed to
cobalt ions showed only
slight or no change in
intracellular cobalt
compared to baseline
levels.
Concentration-dependent effect
(greater effect for NP than ions)
No data
Co NP induced strong
concentration-dependent
increase in ROS, cobalt
ions induced less ROS
and was concentration
independent.
NP readily taken up by
cells and stored in
vacuoles. Proinflammatory activation
after exposure to Co NP
was attributed to
intercellular release of
cobalt ions.
ROS
Cellular uptake
MP = microparticles (diameter > 100 nm), NP = nanoparticles (diameter < 100 nm)
a
Genotoxicity also includes data for related effects (e.g., cell transformation assay) that do not necessarily measure a specific genotoxic endpoint
This draft document should not be construed to represent final NTP determination or policy
111
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Ortega et al. (2014) reported that although cobalt ions were more cytotoxic than poorly soluble
Co3O4 particles, human lung cells exposed to the IC25 (inhibitory concentration at which the ATP
content was reduced by 25% compared to non-exposed cells) of cobalt chloride (2.9 µg/mL) or
Co3O4 (50 µg/mL) had similar intracellular concentrations of solubilized cobalt (6.5 fg/cell for
Co3O4 compared to 5.4 fg/cell for cobalt chloride). Smith et al. (2014) also reported that at
intracellular cobalt concentrations less than 1,000 µM, the cytotoxic effects of cobalt chloride
and CoO to human lung fibroblasts were similar while cobalt chloride was more cytotoxic than
CoO at intracellular concentrations greater than 1,000 µM. Horie et al. (2012) studied a variety
of metal oxide nanoparticles and concluded that cellular influences (cell viability and oxidative
stress) of metal oxide nanoparticles were most dependent on metal ion release (i.e., effects were
greater for soluble particles compared to insoluble particles). In addition, Auffan et al. (2009)
reported that chemically stable nanoparticles did not have significant cellular toxicity while
nanoparticles that could be oxidized, reduced, or dissolved were cytotoxic and genotoxic. Thus,
the available data indicate that intracellular cobalt ions are the primary toxic form and it is likely
that the mode of action for systemic toxicity is related to cobalt ions (Ortega et al. 2014, Smith et
al. 2014, Paustenbach et al. 2013, Simonsen et al. 2012).
Because of similar physical/chemical properties, cobalt ions compete with essential divalent
metal ions (e.g., calcium, copper, zinc, iron, manganese, and magnesium) for absorption, specific
receptor activation, and ion channel transport (Paustenbach et al. 2013). For example, cobalt
absorption is increased in humans and animals with iron deficiency suggesting that these metals
share a common uptake mechanism (Thomson et al. 1971). Further, cobalt ions have the same
size and charge as zinc ions; therefore, both ions bind to the same types of ligands (e.g., oxygen,
nitrogen, and sulfur groups of biomolecules) (Beyersmann and Hartwig 2008). The
bioavailability of cobalt ions in vivo is limited because of extensive binding (90% to 95%) to
serum proteins (e.g., albumin, α2-macroglobulin) and formation of insoluble complexes in the
presence of physiological concentrations of phosphates (Paustenbach et al. 2013, Simonsen et al.
2012). Thus, the concentration of free, ionized cobalt in serum is about 5% to 12% of the total
cobalt concentration (Simonsen et al. 2012). However, Heath et al. (1969) demonstrated that
myoblasts exposed to cobalt-bound protein complexes (primarily globulin and albumin), but not
to cobalt chloride, developed cytological alterations (e.g., enlarged hyperchromatic nucleoli,
chromocenters, and nuclei) in actively growing cultures that were similar to those seen in premalignant myoblasts in vivo. In contrast, myoblasts exposed to cobalt chloride were either killed
or showed no cytological abnormalities when exposed to sublethal concentrations.
Differences in toxicity reported for cobalt particles and ions may be partially explained by
differences in cellular uptake mechanisms. Cobalt ions first saturate binding sites in the
extracellular milieu and on cell surfaces and, after saturation, are actively transported inside the
cell via metal ion transport systems such as calcium channels or divalent metal ion transporters
(Sabbioni et al. 2014, Smith et al. 2014, Simonsen et al. 2012, Garrick et al. 2003). However,
current knowledge of the molecular mechanisms of cobalt ion-specific transporters is very
limited (Guskov and Eshaghi 2012). In contrast, particulate cobalt is transported into cells by
phagocytosis/endocytosis. However, nanoparticles are not as readily phagocytized by alveolar
macrophages as larger particles and also may enter the systemic circulation by penetrating
through the alveolar membrane (Mo et al. 2008).
112
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Studies with low-solubility cobalt oxide (CoO or Co3O4) particles show that these particles
readily enter cells through endocytosis via a clathrin-mediated pathway (called a Trojan-horse
type mechanism) and are partially solubilized in the low pH environment within the lysosomes
(Ortega et al. 2014, Smith et al. 2014, Papis et al. 2009, Limbach et al. 2007). Although the
intracellular solubilized cobalt content was small compared to the intracellular particulate
content, the solubilized fraction was shown to be responsible for the overall toxicity to human
lung cells (Ortega et al. 2014, Smith et al. 2014).
Endocytosis of Co3O4 particles was a more efficient uptake pathway compared to the specific
transport or ionic pumps involved with uptake of cobalt ions. These studies also demonstrated
that concentrations of extracellular solubilized cobalt were too low to induce cytotoxicity and
that particle-to-cell contact was necessary to generate high intracellular cobalt levels. Further,
cobalt particles taken up by lung cells can lead to long-term intracellular release of toxic metal
ions. Similarly, cobalt metal nanoparticles are internalized by phagocytosis and endocytosis and
rapidly spread to the cytosol, cellular organelles, and nucleus where they release cobalt ions
(Sabbioni et al. 2014, Ponti et al. 2009). However, one study reported that the toxic effects of
aggregated cobalt metal nanoparticles in vitro were attributed to extracellular release of cobalt
ions from particle dissolution (Horev-Azaria et al. 2011) while another study reported that
extracellular release of cobalt ions had no effect on cell viability (Nyga et al. 2015).
Sabbioni et al. (2014) also reported that the intracellular distribution of cobalt in Balb/3T3 cells
was different following exposure to cobalt nanoparticles compared to cobalt ions. Cells exposed
to cobalt nanoparticles had a higher nuclear fraction and a lower cytosolic fraction than cells
exposed to cobalt ions. The amount of cobalt bound to DNA was significantly greater in cells
exposed to cobalt microparticles than nanoparticles but was the lowest in cells exposed to cobalt
ions (tested concentrations were 10 and 100 µM for 4 hours). Intracellular distribution studies in
primary rhabdomyosarcoma induced by intramuscular injection of metallic cobalt also reported
that most of the total cellular content of cobalt was associated with the nuclear fraction and was
bound by components of the nucleoplasm, chromatin, and nucleoli (Webb et al. 1972, Heath and
Webb 1967).
The in vivo toxicity and carcinogenicity of soluble cobalt sulfate heptahydrate and cobalt metal
particles from the NTP (2014b, 1998) bioassays were recently compared (Behl et al. 2015).
Contrary to expectations, the data indicated that cobalt metal was more toxic and carcinogenic
than the cobalt salt based on the incidence and spectrum of lung neoplasms and extent of
systemic lesions. However, the findings supported the possibility of a common underlying
mechanism of cobalt toxicity irrespective of the form of cobalt exposure based on the following:
(1) common sites of carcinogenicity (lung and adrenal gland) and a similar spectrum of
nonneoplastic, inflammatory, fibrotic and proliferative lesions in the upper respiratory tract
following subchronic and chronic exposure; (2) similar mutation spectrum in the K-ras oncogene
in lung tumors; (3) toxicity in common extra-pulmonary sites; and (4) similar clinical findings.
Possible explanations for the reported differences between cobalt particles and ions may involve
a synergistic effect between the particles and the transition metal on reactive oxygen species
(ROS) release and/or differences in intracellular cobalt accumulation and distribution (Sabbioni
et al. 2014, Smith et al. 2014, Peters et al. 2007).
This draft document should not be construed to represent final NTP determination or policy
113
06/05/15
6.2
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Proposed modes of action of cobalt carcinogenicity
Similar cytotoxic, genotoxic, and carcinogenic effects have been described for soluble and
particulate forms of cobalt. Three major mechanisms have been identified that are applicable for
the majority of carcinogenic metal compounds (Angelé-Martínez et al. 2014, Koedrith and Seo
2011, Beyersmann and Hartwig 2008). These include (1) oxidative stress, (2) DNA repair
modulation, and (3) disturbances of signal transduction pathways that affect cell growth and
differentiation. Modes of action most likely involved in cobalt-induced carcinogenesis are
consistent with these general mechanisms and include: (1) genotoxicity and inhibition of DNA
repair, (2) induction of reactive oxygen species (ROS) and oxidative stress, and (3) induction of
hypoxia-like responses by activating hypoxia-inducible factor 1 (HIF-1) (see Figure 6-1) (Smith
et al. 2014, Green et al. 2013, Magaye et al. 2012, Simonsen et al. 2012, Simonsen et al. 2011,
De Boeck et al. 2003a, Lison et al. 2001). In addition, these modes of action also affect cell
signaling pathways and gene expression that likely contribute to neoplastic development and
progression (Davidson et al. 2015, Nyga et al. 2015, Verstraelen et al. 2014, Permenter et al.
2013, Malard et al. 2007). Experimental evidence for these modes of action is briefly reviewed
below.
114
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
Figure 6-1. Proposed modes-of-action of cobalt carcinogenicity.
(adapted from Beyersmann and Hartwig 2008, De Boeck et al. 2003a)
This draft document should not be construed to represent final NTP determination or policy
115
06/05/15
6.2.1
Peer-Review Draft RoC Monograph on Cobalt
Genotoxicity, inhibition of DNA repair, and related effects
This section addresses genotoxicity and related biological adverse effects (e.g., cell
transformation, cell-cycle arrest) that are possibly relevant to the mode of action of cobaltinduced carcinogenicity. Genotoxicity (e.g., DNA reactivity, mutagenicity, chromosomal
damage, enzyme-mediated effects on DNA damage or repair) are well recognized as key events
associated with carcinogenesis (Guyton et al. 2009).
The genotoxicity and related effects for cobalt metal and soluble and insoluble cobalt compounds
are reviewed in Appendix E and summarized here (see Table 6-2). Increases in gene mutations,
DNA strand breaks, sister chromatid exchange, micronuclei, aneuploidy, chromosomal
aberrations, DNA-protein crosslinks, inhibition of DNA repair, and cell transformation were
reported in mammalian cells in vitro following exposure to cobalt, but cobalt compounds were
mostly non-mutagenic in bacterial assays. Effects were reported for a variety of cobalt
compounds, including water-soluble salts (chloride, sulfate, nitrate), poorly water-soluble cobalt
compounds (oxide, sulfide, metal, nanoparticles) and a water-soluble organic cobalt compound
(acetate). Although the number of available in vivo studies was limited, they indicated that cobalt
chloride induced genotoxic effects including aneuploidy in the bone marrow and testes of male
hamsters and chromosomal damage and micronucleus formation in mouse bone marrow; cobalt
acetate caused oxidative DNA damage in rat kidney, liver, and lung. Dose-dependent responses
were reported in some of these studies, supporting the evidence for a genotoxic effect in vivo.
More recent in vitro studies are consistent with the earlier data and show that cobalt ions and
particles induce genotoxic effects in human and animal cells, but they also compare effects and
relative potency of cobalt ions and particles (Table 6-1) (Smith et al. 2014, Alarifi et al. 2013,
Patel et al. 2012, Ponti et al. 2009, Colognato et al. 2008). Smith et al. (2014) compared the
effect of CoO particles with cobalt chloride and reported similar genotoxic effects (primarily
chromatid lesions); however, particle-to-cell contact was required to induce genotoxicity from
CoO. Soluble cobalt also induced cell-cycle arrest at a much lower intracellular cobalt
concentration than CoO. Alarifi et al. (2013) compared Co3O4 nanoparticles and cobalt chloride
and reported that both forms caused DNA damage in human HepG2 cells but the nanoparticles
were more potent. Two studies investigated the genetic effects of metallic cobalt nanoparticles in
Balb/3T3 mouse fibroblast (Patel et al. 2012) and human leukocytes (Colognato et al. 2008).
Cobalt nanoparticles induced DNA strand breaks, micronuclei, and cell transformation in mouse
fibroblast and DNA damage in human leukocytes. Cobalt ions had no effect in human leukocytes
but induced DNA damage in mouse fibroblasts. Ponti et al. (2009) reported that cobalt chloride
induced double-strand breaks in human lung epithelial cells and that the effects were increased
with co-exposure to nickel chloride.
Evidence for cobalt-induced inhibition of DNA repair comes from several studies that show
exposure to cobalt enhances the genotoxic effects of some mutagens and that cobalt modifies the
catalytic activity of DNA repair proteins (Beyersmann and Hartwig 2008, IARC 2006,
Beyersmann and Hartwig 1992). It is thought that interaction with DNA repair proteins,
transcription factors, and tumor suppressors may be more relevant for metal-mediated
carcinogenesis than direct binding to DNA (Koedrith and Seo 2011, Beyersmann and Hartwig
2008). Possible mechanisms include substitution of cobalt ions for zinc ions resulting in proteins
with modified catalytic activity (e.g., p53 tumor suppressor protein and zinc finger domains of
116
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
DNA repair proteins) or substitution of cobalt for magnesium in DNA polymerases or
topoisomerases (Beyersmann and Hartwig 2008, Witkiewicz-Kucharczyk and Bal 2006, Baldwin
et al. 2004, Kopera et al. 2004, Asmuss et al. 2000, Hartwig 1998, Kasten et al. 1997, Hartwig et
al. 1991). The DNA binding capacity of p53 protein can be modulated by cobalt(II) ions
(Adámik et al. 2015, Lee et al. 2001, Méplan et al. 2000, Palecek et al. 1999). In addition to cell
cycle arrest and apoptosis, p53 and its downstream genes also regulate DNA excision repair
pathways, including repair of oxidative damage (Smith and Seo 2002). Kasten et al. (1997)
reported that non-cytotoxic doses of cobalt enhanced DNA damage caused by ultraviolet
radiation in human fibroblasts by inhibiting both the incision and polymerization steps of
nucleotide excision repair. Kopera et al. (2004) and Asmuss et al. (2000) showed that cobalt
reduced the DNA-binding ability of xeroderma pigmentosum group A (XPA) protein (a zinc
finger protein involved in nucleotide excision repair). Further, poly(ADP-ribose)polymerase
(PARP), a DNA strand break repair protein also was inhibited by cobalt (Hartwig et al. 2002).
The co-mutagenic effects of cobalt observed in vitro are consistent with one study by Steinhoff
and Mohr (1991) that reported co-carcinogenic effects of cobalt oxide and benzo[a]pyrene for
squamous-cell carcinoma of the lung described in Section 5.2.3.
Genotoxicity assays with cobalt salts and cobalt metal demonstrate a mutagenic potential and at
least two molecular mechanisms seem to apply: (1) a direct effect of cobalt(II) ions to induce
oxidative damage to DNA through a Fenton-like mechanism, and (2) an indirect effect of
cobalt(II) ions through inhibition of repair of DNA damage caused by endogenous events or
induced by other agents (Lison 2015, IARC 2006).
This draft document should not be construed to represent final NTP determination or policy
117
06/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Table 6-2. Summary assessment of genotoxicity and related effects for cobalt compounds
Cobalt
chloride
Endpoint
(Test system)
In
vitro a
In
vivo
Cobalt
sulfate
In
vitro a
In
vivo
Cobalt
nitrate
In
vitro b
In
vivo
Cobalt
oxide
In
vitro b
In
vivo
Cobalt
acetate
In
vitro b
Cobalt
metal
In
vivo
In
vitro a
Cobalt
sulfide
In
vivo
In
vitro b
In
vivo
Cobalt
nanoparticles
In
vitro b
Mutation
Mutation (prokaryotes)
(–)1
Mutation (eukaryotes)
±
(–)1
(–)1
+
+
±
–
Chromosomal damage/cytogenetic effects
Chromosomal
aberrations
+
+
Micronucleus induction
±
+
Recombination
+
Gene conversion
+
–
±
–
+
–
+
+
(+)
Aneuploidy
+
Sister chromatid
exchange
+
+
+
DNA damage and repair
DNA damage/ strand
breaks or bases
+
DNA repair inhibition
+
+
+
+
+
+
+
+
+
Binding/cross-links
DNA-protein crosslinks
+
DNA-protein binding
inhibition
+
+
+
+
+
Other endpoints
Transformation
±
Apoptosis
+
+
+
–
+
+
Sources: IARC (2006) review and additional primary references as described in tables and text.
Positive +, mostly positive evidence (+), mixed results ±, mostly negative evidence (–), and negative –.
118
This draft document should not be construed to represent final NTP determination or policy
+
In
vivo
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
06/05/15
a
Results shown are for –S9; test +S9 was negative.
Results shown are for –S9; not tested with the addition of metabolic activation (S9).
b
This draft document should not be construed to represent final NTP determination or policy
119
This Page Intentionally Left Blank
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6.2.2
6/05/15
Oxidative stress
Reactive oxygen species (ROS) and reactive nitrogen species (RNS) induce oxidative and
nitrative stress and are recognized as key contributors to carcinogenesis (Mates et al. 2010).
Redox-active transition metals (e.g., iron, zinc, copper, chromium, cobalt, nickel, manganese)
have been shown to produce oxidative stress through redox reactions in vivo and in mammalian
cells in vitro (Jomova and Valko 2011, Koedrith and Seo 2011, Beyersmann and Hartwig 2008,
Valko et al. 2006, Valko et al. 2005, Kasprzak 2002). Oxidative stress has been demonstrated to
be one of the principle injury mechanisms through which metal and metal oxide nanoparticles
induce adverse health effects (Zhang et al. 2012b). In addition, cobalt nanoparticles that are
translocated from the lungs to the blood may directly or indirectly activate peripheral blood
neutrophils to release ROS, RNS, and pro-inflammatory cytokines (e.g., IL-1, IL-6, IL-12, MIP2, and TNF-α) (Mo et al. 2008). Excessive or inappropriate neutrophil activation is recognized
as a potential cause of tissue damage. Increased formation of reactive ROS/RNS can overwhelm
body antioxidant defenses leading to oxidative stress and damage to lipids, proteins, and DNA
(Romero et al. 2014, Jomova and Valko 2011, Petit et al. 2005, Valko et al. 2005). Petit et al.
(2005) reported that cobalt ions induced a time- and dose-dependent protein oxidation in human
U937 macrophages that was inhibited by glutathione. In addition to generating DNA damage,
ROS also activate redox-sensitive transcription factors (e.g., NF-κB, AP1, p53) (Beyersmann
and Hartwig 2008, Valko et al. 2006, Valko et al. 2005). These transcription factors have been
linked to carcinogenesis because of their role in regulating DNA repair, inflammation, cell
proliferation, differentiation, angiogenesis, and apoptosis. Thus, depending on the dose and the
extent and timing of interference, ROS may initiate tumor development by mutagenesis and/or
promote tumor growth by dysregulation of cell growth and proliferation.
Both cobalt ions and cobalt metal can catalyze the formation of ROS in vivo and in vitro
(Chattopadhyay et al. 2015, Annangi et al. 2014, Scharf et al. 2014, Alarifi et al. 2013, Patel et
al. 2012, Papis et al. 2009, Qiao et al. 2009, Kotake-Nara and Saida 2007, Limbach et al. 2007,
Peters et al. 2007, Dick et al. 2003, Pourahmad et al. 2003, Zou et al. 2001, Kawanishi et al.
1994, Hanna et al. 1992, Lewis et al. 1992, 1991, Kadiiska et al. 1989, Kawanishi et al. 1989,
Moorhouse et al. 1985). Cobalt sulfate heptahydrate and cobalt(II) acetate (PubChem 2015) were
strongly active in the antioxidant response element signaling pathway (Nrf2/ARE assay) in
human hepatocellular carcinoma (HepG2) cells (Shukla et al. 2012). Cobalt chloride-induced
apoptosis in rat pheochromocytoma (PC12) cells was attributed to ROS formation (Pulido and
Parrish 2003, Zou et al. 2001). Treatment with antioxidants suppressed ROS formation and
blocked apoptosis. Annangi et al. (2014) reported that oxidative stress exacerbated the
acquisition of a cancer-like phenotype as indicated by greater sensitivity of Ogg knockout mouse
embryonic fibroblasts compared to wild-type cells. Scharf et al. (2014) conducted a proteomic
analysis of periprosthetic tissues collected from joint replacement patients during surgery and
reported that cobalt ions induced oxidative damage to proteins involved in the cellular redox
system, metabolism, molecular transport, cellular motility, cell signaling, and organelle function.
Dick et al. (2003) reported evidence for a role of ROS in the toxic and inflammatory effects in
rat lung following intratracheal instillation of Co3O4, and Lewis et al. (1992, 1991) reported
evidence of oxidative stress in hamster lung following exposure to cobalt ions in vivo and in
vitro. Evidence of oxidative stress included decreased levels of reduced glutathione, increased
levels of oxidized glutathione, and increased activity of the pentose phosphate pathway.
Simultaneous incubation with hydrogen peroxide potentiated cobalt-induced increases in levels
This draft document should not be construed to represent final NTP determination or policy
121
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
of oxidized glutathione and pentose phosphate pathway activity. Although the data suggested
that oxidation of glutathione occurred as an early event in cobalt-induced lung toxicity, the data
did not indicate that glutathione oxidation related directly to the observed toxicity. Thus,
oxidative effects that occur at sites other than the glutathione system may mediate cobalt
toxicity. Kasprzak et al. (1994) also reported oxidative damage to DNA in the liver, kidney, and
lung of rats injected with cobalt ions.
Two studies, using different cobalt forms, evaluated K-ras mutations (Figure 6-2) in cobaltinduced lung neoplasms of B6C3F1 mice (cobalt metal and cobalt sulfate heptahydrate) or
F344/NTac rats (cobalt metal only). Rodents were exposed by inhalation (Hong et al. 2015, NTP
2014a, 1998). Both studies found a higher frequency of G to T transversions in codon 12 of the
K-ras gene in cobalt-induced neoplasms compared to spontaneous lung neoplasms from
historical control or other laboratory control rodents. In contrast, the predominant type of K-ras
mutation observed in spontaneous lung tumors from historical control mice was G to A
transitions. No K-ras mutations were observed in spontaneous lung tumors in the concurrent or
historical control rats. K-ras G to T transversion mutations are associated with the production of
8-hydroxydeoxyguanosine adducts that result from oxidative damage (Itsara et al. 2014, Klaunig
et al. 2010) and are consistent with the mutation pattern in bacteria (i.e., positive in strains
detecting mutational events at G:C base pairs). G to T are also the most common type of
mutation observed in human lung tumors in the p53 gene (Harty et al. 1996). These studies
suggest that suggest that oxidative damage may play a role in cobalt-mediated lung
tumorigenicity.
Figure 6-2. K-ras mutations in lung tumors from cobalt-exposed and non-exposed rodents
Source: (Hong et al. 2015, NTP 2014a, 1998).
HC = historical control, M = mouse, R = rat
Frequency of K-ras mutations from lung tumors in mice or rats exposed to cobalt metal or cobalt sulfate and spontaneous tumors.
Total K-ras is the incidences of any Kras mutation detected in all samples and includes mutations in codon 12, 13, and 61. G to T
and G to A is the frequency of these specific mutations occurring in codon 12 only. i.e., the total number of K-ras mutations in
codon 12 is the denominator.
122
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Direct interactions between cobalt metal or ions and oxygen or lipids can generate ROC. High
concentrations (10 mg/mL) of aqueous suspensions of Co(0) metal particles can react with
dissolved oxygen to generate hydrogen peroxide and hydroxyl radicals in the presence of
superoxide dismutase (SOD) as illustrated below (reactions 1-3) (Lee et al. 2012, Jomova and
Valko 2011, Leonard et al. 1998). The hydroxyl radical was not generated when catalase, a
hydrogen peroxide scavenger, was added. Cobalt(II) ions alone did not generate significant
amounts of hydroxyl radicals from hydrogen peroxide except when bound to certain endogenous
chelators such as glutathione and anserine (reaction 4) (Leonard et al. 1998, Mao et al. 1996, Shi
et al. 1993). Glutathione and anserine normally function as antioxidants; however, these data
suggest that a cobalt(II)-mediated switch to pro-oxidants may occur and cause cellular damage
(Valko et al. 2005). Cobalt(II) ions also are capable of reacting with lipid hydroperoxides to
generate free radicals in the presence of proper chelating agents (Shi et al. 1993). Hydroxyl
radicals and lipid hydroperoxide-derived free radicals are considered important intermediates in
oxidative stress-induced genetic damage and as mediators of tumor initiation and promotion
(Barrera 2012, Shi et al. 1993, Vaca et al. 1988). Thus, under certain conditions, both cobalt
metal and cobalt ions are capable of generating ROS through Fenton-like reactions (reactions 3
and 4) with the potential to increase oxidative stress and cellular injury through DNA damage,
protein modification, induction of oncogene expression, and nuclear transcription factor
activation.
One argument against the oxidative-stress hypothesis of metal-induced carcinogenesis is that
high, cytotoxic doses of metals (e.g., mM range) are often required to induce oxidative damage
while much lower doses induce tumors (Paustenbach et al. 2013, Beyersmann and Hartwig
2008). However, as mentioned above, G to T transversions in mouse and rat lung tumors induced
by cobalt sulfate and/or cobalt metal are characteristic of oxidative damage. Further, sub-toxic
doses of cobalt nanoparticles induced oxidative stress and cell transformation in mouse embryo
fibroblasts (Annangi et al. 2014, Sighinolfi et al. 2014) and oxidative stress and DNA damage in
human lung epithelial (A549) cells (Wan et al. 2012). It has been suggested that oxidative stress
is not the sole cause of cobalt-induced carcinogenicity but may contribute in a potentiating
manner (Beyersmann and Hartwig 2008).
6.2.3
Hypoxia mimicry and HIF-1 stabilization
HIF-1 is a heterodimer composed of HIF-1α and HIF-1β subunits and is the key mediator of
hypoxia response (Davidson et al. 2015, Galanis et al. 2008, Salnikow et al. 2004). The β
subunit, also known as aryl hydrocarbon receptor nuclear translocator (ARNT) is constitutively
expressed while HIF-1α is oxygen sensitive. HIF-1 overexpression and enhanced transcriptional
activity are linked to cancer initiation and progression. There is strong experimental support that
This draft document should not be construed to represent final NTP determination or policy
123
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
HIF-1 activation is involved in cobalt-induced carcinogenesis. Cobalt metal particles, cobalt
chloride, and cobalt sulfate heptahydrate promote a hypoxia-like state in vivo and in vitro, even
with normal molecular oxygen pressure, by stabilizing HIF-1α (Nyga et al. 2015, Galán-Cobo et
al. 2013, Gao et al. 2013, Saini et al. 2010b, Saini et al. 2010a, Galanis et al. 2009, Qiao et al.
2009, Xia et al 2009, Beyersmann and Hartwig 2008, Maxwell and Salnikow 2004). Further,
Wang and Semenza (1995) demonstrated that HIF-1 induction either from hypoxia or cobalt
chloride treatment was indistinguishable with respect to DNA binding specificity and contacts
with target DNA sequences.
Evidence for cobalt-induced HIF-1 stabilization has been demonstrated in several human cell
lines, including cancer cell lines (Fu et al. 2009, Ardyanto et al. 2008, Wang and Semenza
1995). Cobalt chloride-induced hypoxia also increased the invasiveness of one primary breast
cancer cell line (Fu et al. 2009). Under normal oxygen conditions, the iron-containing oxygen
sensing enzymes (oxygenases) hydroxylate specific proline and asparagine residues in the HIF1α subunits (Maxwell and Salnikow 2004). Hydroxylated HIF-1α binds to a multiprotein
complex that contains the VHL tumor suppressor. VHL acts as part of an ubiquitin ligase
complex resulting in rapid ubiquitination and proteolysis of HIF-1α. Under hypoxic conditions,
HIF-1α subunits are not hydroxylated, and consequently the protein is stabilized and translocates
to the nucleus where it binds with a HIF-1β subunit. The response to hypoxia includes increased
red blood cell production, blood vessel growth and increased blood supply to tissues, and
increased anaerobic metabolism. Cobalt affects the function of several genes and enzymes
responsible for posttranslational modification of HIF-1α such as prolyl hydroxylases and VHL
(Davidson et al. 2015). Possible mechanisms by which cobalt ions activate HIF-1 include
replacing iron in the regulatory oxygenases or depleting intracellular ascorbate (a cofactor for
prolyl hydroxylase activity), thus, deactivating these enzymes (Davidson et al. 2015, Qiao et al.
2009, Maxwell and Salnikow 2004, Salnikow et al. 2004). Oxidative stress has also been
investigated as a possible mechanism of cobalt-induced HIF activation; however, Salnikow et al.
(2000) showed that activation of HIF-1-dependent genes was independent from ROS formation.
Nyga et al. (2015) also reported evidence that HIF-1α stabilization in human macrophages
treated with cobalt metal nanoparticles or cobalt ions occurred via an ROS-independent pathway.
HIF-1α is present in almost all human and animal cells and its activation has a central role in the
transcriptional regulation of more than 100 hypoxia-responsive genes (including genes encoding
for multiple angiogenic growth factors (e.g., VEGF), erythropoietin synthesis, endothelin,
glucose transporters, inflammatory factors, and regulation of apoptosis and cell proliferation)
that allow for cell survival at low oxygen pressure (Gao et al. 2013, Simonsen et al. 2012, Saini
et al. 2010b, Saini et al. 2010a, Greim et al. 2009, Beyersmann and Hartwig 2008, Wang and
Semenza 1995). The evidence suggests that HIF-1α is a major regulator of the adaptation of
cancer cells to hypoxia and may contribute to tumor development and progression by decreasing
both repair and removal of mutated cells, selecting for cells with genetic instability, reducing p53
transcriptional activity, evading growth arrest checkpoints, and inducing apoptosis resistance
(Greim et al. 2009, Ardyanto et al. 2008, Hammond and Giaccia 2005, Maxwell and Salnikow
2004, Lee et al. 2001). HIF-1α overexpression, stabilization and transcriptional activation is
found in more than 70% of human cancers (e.g., breast, ovarian, cervical, prostate, brain, lung,
head and neck) and is associated with poor clinical outcomes (Cheng et al. 2013, Galanis et al.
2009, Galanis et al. 2008, Maxwell and Salnikow 2004, Paul et al. 2004). Greim et al. (2009)
also identified hypoxia and HIF activation as a relevant mechanism for pheochromocytomas in
124
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
rats. Further evidence for a role of HIF-1 in cancer is as follows: (1) enhanced glycolytic and
angiogenic activities are hallmarks of many tumors and are consequences of HIF-1 activation,
(2) immunolabelling for HIF-1α subunits confirms there is a common activation in solid tumors,
(3) genetic studies comparing tumor growth with and without HIF-1 have generally shown that
tumors without specific HIF subunits have decreased vascularization and growth, (4) a number
of pathways implicated in cancer progression increase activation of the HIF-1 pathway in
normoxia and hypoxia, and (5) as described above, the VHL tumor suppressor protein is required
to regulate HIF-1 (Maxwell and Salnikow 2004). VHL loss of function results in constitutive
HIF activation and an increased risk of developing cancer.
6.2.4
Cell signaling and gene expression modulation
Cell signaling pathways control the expression of numerous genes that play important roles in
carcinogenesis and many of these pathways are known targets of metal toxicity (Broberg et al.
2015, Davidson et al. 2015). Signaling pathways, receptors and transcription factors affected by
cobalt include MAPKs, HIF-1, p53, AP-1, VEGF, P13K/Akt, and NFκB (Davidson et al. 2015,
Lee et al. 2012, Mates et al. 2010, Leonard et al. 2004). Cobalt-mediated effects in humans and
animal cell lines, which may be direct by interaction with proteins, or indirect through formation
of ROS, are briefly reviewed here.
Human A549 lung adenocarcinoma cells exposed to cobalt chloride overexpressed the N-myc
downstream regulated gene 1 (NDRG1/Cap43) (Salnikow et al. 2000). Increased expression of
NDRG1/Cap43 was reported in tumors and serum of lung cancer patients compared to adjacent
normal tissues and may be predictive of tumor angiogenesis and poor prognosis (Azuma et al.
2012, Wang et al. 2012).
Malard et al. (2007) also reported that A549 cells exposed to cobalt chloride differentially
expressed 85 genes including potential cobalt carriers, tumor suppressors, transcription factors,
and genes linked to stress response. Most of these had never been described as related to cobalt
stress and only 7 of 85 genes matched HIF-1 target genes. In another study, cobalt oxide
nanoparticles at non-cytotoxic concentrations induced mainly downregulation of gene
transcription in A549 and bronchial BEAS-2B cells (Verstraelen et al. 2014). BEAS-2B cells
were more sensitive than A549 cells having higher numbers of differentially expressed
transcripts at a 10-fold lower concentration. Two distinct clusters of upregulated genes were
observed in both cell lines that were associated with metabolic processes. Between 1% and 14%
of the differentially expressed transcripts encoded markers involved in immune processes. Cobalt
nanoparticles and ions also induced a time-dependent increase in HIF-target genes and
expression of proinflammatory cytokines in the U937 human monocytic cell line (Nyga et al.
2015).
Permenter et al. (2013) investigated gene expression and intracellular protein abundance in two
rat liver cell lines exposed to cobalt chloride. Many genes, proteins, and pathways were
modulated, which were mainly due to induction of a hypoxia-like response and oxidative stress.
These data were consistent with gene expression profiling in human hepatocellular carcinoma
(Hep3B) cells exposed to cobalt chloride (Vengellur et al. 2005).
This draft document should not be construed to represent final NTP determination or policy
125
6/05/15
6.3
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Cobalt tissue levels from patients with lung and other cancers
Several publications were identified that measured trace metals (such as heavy metals and
essential metals) in tissue (such as tumor of different stages or normal tissue) or surrogates (e.g.,
hair, nails, blood) from cancer patients with a referent group (e.g., healthy humans, other
diseases) or referent tissue (e.g., non-tumor from the same or different subjects) (see Appendix
B). For most studies, the source of the exposure was unknown. Overall, several studies found
statistically significant higher levels of cobalt in surrogate tissues (hair, nails, urine, or serum)
from patients with several different types of cancer including all cancers (Pasha et al. 2007),
cancer of the lung (Qayyum and Shah 2014, Benderli Cihan and Öztürk Yildirim 2011), larynx
(Collecchi et al. 1986), liver (Yin 1990), or breast (Benderli Cihan et al. 2011) compared to
healthy controls. However, except for lung cancer, there was only one study per specific cancer
site. None of the studies were able to distinguish whether metal levels could be a cause of cancer
or whether the cancer process itself affects metal balances and there were several limitations
such as co-exposure to other metals and limited methods to select cases and controls, and the
source of the exposure is unknown.
6.4
Synthesis
Cobalt metal and several cobalt compounds induce similar carcinogenic effects in experimental
animals. The mechanisms of cobalt-induced neoplasms are not completely understood but the
available data provide strong support that intracellular cobalt ions are the principle toxic entity.
Cobalt ions are actively transported inside the cell via metal ion transport systems while cobalt
particles with low solubility are readily taken up by cells via endocytosis. Once inside the cell,
cobalt particles are partially solubilized at the low pH within lysosomes and release cobalt ions
that can react with DNA, proteins, and lipids. Mechanistic data provide strong support that
inhibition of DNA repair, oxidative stress, and activation of HIF-1α likely contribute to cobaltinduced neoplastic development and progression. All of these mechanisms are relevant to
humans.
126
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
7 Overall Cancer Evaluation and Preliminary Listing
Recommendation
This section brings forward and integrates the evaluations of the human, animal and mechanistic
and other relevant data, applies the RoC listing criteria, and reaches a preliminary listing
recommendation.
Preliminary listing recommendation
“Cobalt and certain cobalt compounds” are reasonably anticipated to be human carcinogens
based on sufficient evidence from studies in experimental animals and supporting mechanistic
data. “Certain” refers to those cobalt compounds – including soluble and poorly water-soluble
cobalt compounds and particles – that can release cobalt ions in vivo, which mechanistic data
indicate are key for cobalt-induced carcinogenicity.
Because mechanistic data are key for the evaluation of cobalt and certain cobalt compounds, that
topic is discussed first (Section 7.1), followed by a discussion of the rationale for grouping cobalt
and certain cobalt compounds as a class (Section 7.2). The scientific data supporting the
conclusion of sufficient evidence of cobalt and certain cobalt compounds from studies in
experimental animals is discussed in Section 7.3, and the conclusions from the cancer studies in
human studies is briefly summarized in Section 7.4.
7.1
Mechanistic and other relevant data
Although the mechanisms of cobalt-induced carcinogenicity are not completely understood,
three biologically plausible modes-of-action have been identified and were reviewed in Section
6. These include genotoxicity and inhibition of DNA repair, ROS and oxidative damage, and
stabilization of HIF-1α. Cobalt ions can replace zinc ions in the zinc finger domains of DNA
repair proteins, thus altering their catalytic activity and in vitro assays consistently show
genotoxic effects in mammalian cells exposed to a wide range of cobalt compounds. Cobalt is a
redox-active transition metal and in vitro studies show that cobalt particles and ions can induce
ROS in mammalian cells with cobalt metal and cobalt oxide particles having a greater effect than
ions. Evidence of oxidative stress and oxidative damage also were shown in in vivo studies.
Finally, HIF-1α stabilization is well established for cobalt. Although most studies used cobalt
chloride to promote a hypoxia-like state, cobalt metal nanoparticles were also shown to have this
effect. HIF-1α plays a central role in the transcriptional regulation of more than 100 hypoxiaresponsive genes and is a major regulator of the adaptation of cancer cells to hypoxia. Although
there were some differences in the degree of toxicity or biological response among cobalt metal
particles, cobalt oxide particles, and cobalt ions the modes of action are relevant for all of these
cobalt forms.
7.2
Cobalt and certain cobalt compounds as a class
Chemical grouping describes a general approach for considering more than one chemical at the
same time for hazard assessment or regulatory purposes. Chemicals whose physicochemical
and/or toxicological properties are likely to be similar or follow a consistent pattern, usually as a
result of structural similarity, may be considered as a group, or category of substances (OECD
This draft document should not be construed to represent final NTP determination or policy
127
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
2014, ECHA 2009). One of the primary advantages of grouping is that every chemical within the
group does not necessarily require testing for every endpoint. Where scientifically justifiable,
chemicals and endpoints that have been tested can be used to fill in the data gaps for the untested
chemicals and endpoints. Obviously, only a limited number of cobalt compounds have been
tested for one or more of the endpoints evaluated in this monograph. Therefore, a group
approach is proposed and the following sections are based on data reviewed in the previous
sections of this document that are relevant to the proposed group listing.
Mechanistic data informed the approach for grouping cobalt and certain cobalt compounds as a
class. The key events involve cellular uptake of cobalt, intracellular release of cobalt ions from
particles, intracellular concentrations and distribution, immediate and downstream molecular
effects (discussed below and illustrated in Figure 6-1), and tumor formation. Thus,
physicochemical properties, toxicokinetics, mechanistic data and other relevant data were used to
identify and compare the chemical and biological properties and events that were relevant to
cobalt-induced carcinogenicity to determine if a group listing for cobalt and certain cobalt
compounds was warranted. These endpoints are compared in Table 7-1 for several cobalt
compounds in Section 7.2.3 and discussed below.
In addition to the mechanistic data, other data relevant for chemical grouping include the
following:
•
Physicochemical properties and toxicokinetics (Section 7.2.1);
•
Toxicological effects related to a common functional group (i.e., the cobalt ion) (Section
7.2.2); and
•
Overall synthesis (Section 7.2.3).
7.2.1
Physicochemical properties and toxicokinetics
Physicochemical properties and toxicokinetic data for cobalt metal and various cobalt
compounds were presented in Sections 1 and 3. Solubility, particle size, bioavailability, and
cellular uptake and retention affect toxicity. These data show the following general rank order
for aqueous solubility: cobalt(II) salts > cobalt metal > cobalt oxides. Bioaccessibility, defined as
the availability of a metal for absorption when dissolved in artificial body fluids, is often used as
an in vitro surrogate for bioavailability testing (Stopford et al. 2003). Bioaccessibility
measurements showed the same general rank order as aqueous solubility at near neutral pH but,
in acidic solutions associated with lysosomes (pH 4.5) or gastric fluid (pH 1.5), bioaccessibility
was 100% or near 100% for cobalt metal and all cobalt compounds tested including watersoluble and poorly soluble compounds indicating that they release cobalt ions in solution.
As discussed in Section 3, a number of factors affect cobalt absorption. This is reflected by the
fact that absorption of cobalt compounds following oral exposure varies widely but soluble
forms are better absorbed than insoluble forms. Inhalation studies also indicate better absorption
and shorter retention in the respiratory tract of soluble forms compared to insoluble forms. Thus,
cobalt particles with low solubility (e.g., cobalt oxides) are retained in the lungs for long periods
and represent a continuing source of exposure. Although cobalt metal has low aqueous solubility,
NTP’s chronic inhalation study showed that lung clearance in rats and mice was similar to that
observed for soluble cobalt sulfate heptahydrate. Cobalt concentrations and tissue burdens
128
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
increased with increasing exposure concentrations in all tissues examined, indicating systemic
exposure; however, normalized tissue burdens increased only in the liver.
Although soluble cobalt compounds are better absorbed, cellular uptake mechanisms for particles
also are important (see Section 6.1). Thus, cellular uptake of poorly soluble cobalt particles via
endocytosis/phagocytosis can result in intracellular dissolution within the lysosomes and release
of cobalt ions. In vitro studies of cobalt metal and cobalt oxide particles generally show that
intracellular cobalt ion release is responsible for toxicity as opposed to extracellular dissolution.
These studies demonstrated that direct particle contact with the cultured cells was required for
cellular uptake, intracellular ion release and toxicity, while cells that were exposed only to
extracellular ions dissolved from the particles were not affected. In contrast, cobalt ions readily
form complexes with proteins and low molecular weight components and must first saturate
binding sites in the extracellular milieu and on cell surfaces before entering the cell via metal ion
transport systems. Solubility, particle size, and particle surface area also affect elimination from
the body. Elimination of cobalt particles and ions is multiphasic with fast, intermediate, and slow
phases; however, soluble compounds are cleared faster with a smaller fraction of the dose
retained long term.
7.2.2
Toxicological effects and key events
In vivo studies in humans and experimental animals consistently show that cobalt and cobalt
compounds induce a similar spectrum of inflammatory, fibrotic, and proliferative lesions in the
upper respiratory tract. Toxicological effects of cobalt are attributed primarily to the cobalt ion;
however, in vitro studies indicate that direct toxic effects of cobalt particles also contribute.
Relevant toxic effects reviewed in this document include carcinogenicity in humans and
experimental animals, genetic and related effects (in vitro and in vivo), oxidative stress (in vitro
and in vivo), and cytotoxicity (in vitro). Although not completely understood, cellular uptake
mechanisms and intracellular release of cobalt ions and their distribution are important factors.
Cobalt metal and cobalt compounds exhibited similar carcinogenic effects in animals and similar
genotoxic and cytotoxic effects in vitro. Inhalation studies with cobalt sulfate or cobalt metal
primarily induced lung tumors (although tumors distal to the lung were found for cobalt metal)
while injection-site tumors were induced following subcutaneous, intraperitoneal, intramuscular,
or intratracheal administration of various cobalt particles and compounds. In vitro assays show
that cobalt metal and cobalt compounds induce genetic damage and inhibit DNA repair. In vivo
genotoxicity data were mostly conducted with cobalt chloride and were positive for aneuploidy,
micronucleus formation and chromosomal aberrations; cobalt acetate caused DNA damage in the
lung and several other tissues. In vitro cytotoxicity assays were consistent in reporting doserelated effects for cobalt metal particles, cobalt oxide particles, and cobalt ions. In general,
metallic cobalt particles induced cytotoxicity, ROS formation, genotoxicity and carcinogenicity
to a greater extent than cobalt ions while cobalt oxide particles with low solubility were less
cytotoxic than cobalt ions but induced higher levels of ROS (see Table 6-1). Many studies (both
in vitro and in vivo) have reported evidence that cobalt induces oxidative stress, particularly
when complexed with endogenous chelators such as glutathione or anserine. In addition,
mutations in lung tumors induced by cobalt sulfate or cobalt metal included G to T transversions
that are characteristic of oxidative damage.
This draft document should not be construed to represent final NTP determination or policy
129
6/05/15
7.2.3
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Overall synthesis
Several biological endpoints were identified from physicochemical, toxicological, and
mechanistic data for cobalt metal, cobalt chloride, cobalt sulfate, and cobalt oxide (CoO and
Co3O4). These cobalt forms were the most studied and included both soluble and insoluble
forms. The data are synthesized and compared on a semi-quantitative scale in Table 7-1.
Symbols (i.e., −, +, ++, +++) and colors are used to designate the relative effect of the various
cobalt forms (see explanation below). These data provide justification for the proposed group
approach and are consistent with the OECD (2014) and ECHA (2009) guidelines for chemical
grouping. Thus, biological properties of cobalt compounds that are not included in this table may
be inferred by comparing to an analogous cobalt compound within the table.
Table 7-1 also includes particle size (i.e., < 100 nm or > 100 nm) for cobalt metal and cobalt
oxides to compare relative effects of nanoparticles with larger particles. Although many of the in
vitro studies included direct comparisons of endpoints for cobalt ions and cobalt particles, few
studies provided a direct comparison of particles of different sizes. In those cases, determinations
for the particle sizes were made relative to cobalt ions. There also were a few cases where the
available studies were inconsistent. In those cases, it was not possible to determine the relative
effects of the particles or compounds; therefore, the effects were considered equivalent.
However, cell designations in Table 7-1 should only be compared within a given row or, in some
cases, within a column (i.e., bioaccessibility data or animal neoplasms) because the relative
scales vary by endpoint. For example, the cell designation for lung tumors induced by cobalt
sulfate indicates a less potent response when compared to lung tumors induced by cobalt metal
but should not be compared to the cell designation for adrenal tumors induced by cobalt metal.
130
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Table 7-1. Comparison of chemical and biological properties of cobalt metal and cobalt compounds.
Soluble compounds
Cobalt metal
Poorly soluble
compounds
Cobalt oxidea
CoCl2
Endpoint
CoSO4
> 100 nm
< 100 nm
>100 nm
<100 nm
b
Bioacessibility
Alveolar
+++
+++
++
ND
+
ND
Interstitial/serum
+++
+++
++
++
+
+
Lysosome
+++
+++
+++
+++
+++
+++
Gastric
+++
+++
+++
ND
+++
ND
Cellular uptake
+
ND
+++
+++
+++
+++
Cytotoxicity
++
++
+++
+++
+
++
ROS
+
ND
++
++
ND
++
+++
+++
ND
+++
ND
ND
DNA repair inhibition
+
ND
+
ND
ND
ND
Genotoxicity in vitro
+
+
+
+
+
++
Genotoxicity in vivo
+
ND
−
ND
ND
ND
Lung
ND
++
+++
ND
+
ND
Adrenal gland
ND
+
++
ND
ND
ND
Pancreatic islet
ND
−
+
ND
ND
ND
Kidney
ND
−
−
ND
ND
ND
Mononuclear cell leukemia
ND
−
+
ND
ND
ND
c
+
ND
+++
+++
+++
ND
HIF-1α stabilization
Animal neoplasms
Injection site
ND = no data, − = no effect, + = positive effect, ++ = greater positive effect, +++ = greatest positive effect. (Note: the
relative scales vary by endpoint.)
a
Includes CoO and Co3O4.
b
Dissolution in artificial extraction fluids or culture medium after 48 to 72 hours.
c
Includes subcutaneous, intramuscular, intraperitoneal, and intrathoracic injection or implantation studies.
7.3
Evidence of carcinogenicity from studies in experimental animals
There is sufficient evidence for the carcinogenicity of cobalt and certain cobalt compounds
(collectively referred to as cobalt) in experimental animals based on increased incidence of
malignant and/or a combination of malignant and benign neoplasms at several tissue sites in rats
and mice by different routes of exposure. Inhalation exposure to cobalt caused dose-related
increases in the incidence of lung neoplasms (mainly alveolar/bronchiolar adenoma and
carcinoma) in male and female mice and rats, adrenal gland (benign and malignant
pheochromocytoma) in male and female rats, hematopoietic system (mononuclear-cell leukemia)
in female rats, and pancreas (islet-cell adenoma or carcinoma combined) in male rats. (Evidence
is insufficient to differentiate between a direct and indirect cause of adrenal gland neoplasms
from cobalt exposure.) Tumors of the pancreas (islet-cell carcinoma) in female rats and kidney
(adenoma or carcinoma combined) in male rats may have been related to exposure to cobalt
metal. Injection-site tumors (such as sarcoma, histiocytoma, rhabdomyofibrosarcoma, or
This draft document should not be construed to represent final NTP determination or policy
131
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
fibrosarcoma) were observed in rats exposed to different forms of cobalt by parenteral
administration (such as intramuscular, subcutaneous, intraperitoneal injection).
Both lung and injection-site tumors were induced in rodents by different forms of cobalt,
including cobalt metal, and soluble (e.g., cobalt sulfate or cobalt chloride) and poorly soluble
cobalt compounds (cobalt oxide). A comparison of the inhalation studies conducted by NTP of
cobalt metal and cobalt sulfate suggest that cobalt metal was more toxic and carcinogenic at a
similar cobalt concentration as evidenced by the incidence and spectrum of lung neoplasms and
the extent of systemic lesions. This is consistent with mechanistic studies showing that cobalt
metal has a greater effect on ROS than cobalt ions.
7.4
Evidence of carcinogenicity from studies in humans
There is inadequate evidence from studies in humans to evaluate the association between
exposure to cobalt and cancer. While almost all the cohort studies reported approximately a
doubling of the risk of lung cancer from exposure to various cobalt compounds, cobalt exposure
was likely correlated with exposure to other known lung carcinogens, which complicates the
interpretation of the results. Increased risks of esophageal cancer were found in two populationbased case-control studies; however, cobalt exposure was assessed in toenail samples at or after
cancer diagnosis. Thus, it is unclear whether cobalt levels in the toenails reflected exposure to
cobalt preceding cancer or resulted from changes due to tumor formation.
132
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
8 References
1.
Abbracchio MP, Heck JD, Costa M. 1982. The phagocytosis and transforming activity of
crystalline metal sulfide particles are related to their negative surface charge.
Carcinogenesis 3(2): 175-180. (Supported by NIEHS, the U.S. EPA and the Department of
Energy. Authors affiliated with University of Texas, TX. )
2.
Abrams JA, Fields S, Lightdale CJ, Neugut AI. 2008. Racial and ethnic disparities in the
prevalence of Barrett's esophagus among patients who undergo upper endoscopy. Clin
Gastroenterol Hepatol 6(1): 30-34. (Supported by NCI. Authors affiliated with Columbia
University Medical Center, NY.)
3.
Adachi S, Takemoto K, Ohshima S, Shimizu Y, Takahama M. 1991. Metal concentrations
in lung tissue of subjects suffering from lung cancer. Int Arch Occup Environ Health 63(3):
193-197. (Support not reported. Authors affiliated with Saitama Medical School, Japan.)
4.
Adámik M, Bažantová P, Navrátilová L, Polášková A, Pečinka P, Holaňová L, Tichý V,
Brázdová M. 2015. Impact of cadmium, cobalt and nickel on sequence-specific DNA
binding of p63 and p73 in vitro and in cells. Biochem Biophys Res Commun 456(1): 29-34.
(Supported by the IGA VFU Brno, the Czech Science Foundation and IET. Authors
affiliated with Academy of Science of the Czech Republic, Czech Republic; University of
Veterinary and Pharmaceutical Sciences, Czech Republic; University of Ostrava, Czech
Republic.)
5.
Afridi HI, Kazi TG, Kazi NG, Jamali MK, Arain MB, Sirajuddin, Kandhro GA, Shah AQ,
Baig JA. 2009. Evaluation of arsenic, cobalt, copper and manganese in biological Samples
of Steel mill workers by electrothermal atomic absorption Spectrometry. Toxicol Ind
Health 25(1): 59-69. (Supported by the Higher Education Commission, Islamabad,
Pakistan. Authors affiliated with University of Sindh, Pakistan.)
6.
Akbar M, Brewer JM, Grant MH. 2011. Effect of chromium and cobalt ions on primary
human lymphocytes in vitro. J Immunotoxicol 8(2): 140-149. (Supported by EPSRC and
DePuy International Ltd. Authors affiliated with University of Strathclyde, UK; University
of Glasgow, UK.)
7.
Alarifi S, Ali D, Y AO, Ahamed M, Siddiqui MA, Al-Khedhairy AA. 2013. Oxidative
stress contributes to cobalt oxide nanoparticles-induced cytotoxicity and DNA damage in
human hepatocarcinoma cells. Int J Nanomedicine 8: 189-199. (Supported by the King
Saud University. Authors affiliated with King Saud University, Saudi Arabia.)
8.
Alessio L, Dell'Orto A. 1988. Biological monitoring of cobalt. In Biological Monitoring of
Toxic Metals. T.W. C, Friberg L, Nordberg GF, Sager PR, eds. New York, NY: Springer
US.pp. 407-417. (Support not reported. Authors affiliated with University of Brescia,
Italy.)
9.
Alexandersson R, Lidums V. 1979. [Studies on the effects of exposure to cobalt. IV. Cobalt
concentrations in blood and urine as exposure indicators]. Arbete Halsa 8: 2-23. (as cited in
IARC 2006)
This draft document should not be construed to represent final NTP determination or policy
133
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
10.
Alexandersson R. 1988. Blood and urinary concentrations as estimators of cobalt exposure.
Arch Environ Health 43(4): 299-303. (Support not reported. Authors affiliated with
Sundsvall's Hospital, Sweden.)
11.
Alimonti A, Bocca B, Lamazza A, Forte G, Rahimi S, Mattei D, Fiori E, Iacomino M,
Schillaci A, De Masi E, Pino A. 2008. A study on metals content in patients with colorectal
polyps. J Toxicol Environ Health A 71(5): 342-347. (Supported by the LILT, Lega Italiana
per la Lotta contro i Tumori (Italian League against Cancer). Authors affiliated with Italian
National Institute of Health, Italy; Hospital ‘Umberto I', Italy; Hospital ‘San Carlo di
Nancy’, Italy.)
12.
Alipázaga MV, Moreno RG, Linares E, Medeiros MH, Coichev N. 2008. DNA damage by
sulfite autoxidation catalyzed by cobalt complexes. Dalton Trans(41): 5636-5644.
(Supported by the Fundação de Amparo à Pesquisa do Estado de S˜ao Paulo (FAPESP) and
Conselho Nacional de Pesquisa e Desenvolvimento Tecnol´ogico (CNPq) (Brazilian
Agencies))
13.
Allen MJ, Myer BJ, Millett PJ, Rushton N. 1997. The effects of particulate cobalt,
chromium and cobalt-chromium alloy on human osteoblast-like cells in vitro. J Bone Joint
Surg Br 79(3): 475-482. (Support not reported. Authors affiliated with Addenbrooke’s
Hospital, UK.)
14.
Amacher DE, Paillet SC. 1980. Induction of trifluorothymidine-resistant mutants by metal
ions in L5178Y/TK+/- cells. Mutat Res 78(3): 279-288. (as cited in IARC 2006)
15.
Anard D, Kirsch-Volders M, Elhajouji A, Belpaeme K, Lison D. 1997. In vitro genotoxic
effects of hard metal particles assessed by alkaline single cell gel and elution assays.
Carcinogenesis 18(1): 177-184. (Supported by the Federal Services of the Scientific,
Technical and Cultural Affairs of the Prime Minister’s Service (Belgium). Authors
affiliated with Universite Catholique de Louvain, Belgium; Vrije Univesiteit Brussel,
Belgium.)
16.
Andersen O. 1983. Effects of coal combustion products and metal compounds on sister
chromatid exchange (SCE) in a macrophagelike cell line. Environ Health Perspect 47: 239253. (as cited in IARC 2006)
17.
Anderson LA, Cantwell MM, Watson RG, Johnston BT, Murphy SJ, Ferguson HR,
McGuigan J, Comber H, Reynolds JV, Murray LJ. 2009. The association between alcohol
and reflux esophagitis, Barrett's esophagus, and esophageal adenocarcinoma.
Gastroenterology 136(3): 799-805. (Supported by an Ireland-Northern Ireland Cooperation Research Project Grant sponsored by the Northern Ireland Research &
Development Office, by the Health Research Board, Ireland, and by the Ulster Cancer
Foundation. Authors affiliated with Queen’s University, Ireland; Royal Group of Hospitals,
Ireland; Daisy Hill Hospital, Ireland; National Cancer Registry, Ireland; St. James’s
Hospital, Ireland.)
18.
Andre S, Metivier H, Masse R. 1989. An interspecies comparison of the lung clearance of
inhaled monodisperse cobalt oxide particles - Part III: Lung clearance of inhaled cobalt
134
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
oxide particles in baboons. J Aerosol Sci 20(2): 205-217. (Support not reported. Authors
affiliated with Institut de Protection et de Surete Nucleaire, France.)
19.
Angelé-Martínez C, Goodman C, Brumaghim J. 2014. Metal-mediated DNA damage and
cell death: Mechanisms, detection methods, and cellular consequences. Metallomics 6(8):
1358-1381. (Supported by the National Science Foundation. Authors affiliated with
Clemson University, SC.)
20.
Angerer J. 1989. Cobalt. In Biologische Arbeitsstoff- Toleranzwerte (BAT-Werte),
Arbeitsmedizinisch-toxikologische Bergründungen [Biological Occupational Tolerance
Value, Occupational Medical-toxicological Basis]. Henschler D, Lehnert G, eds.
Weinheim: VCH-Verlag. pp. 1-13. (as cited in IARC 1991)
21.
Annangi B, Bach J, Vales G, Rubio L, Marcos R, Hernández A. 2014. Long-term
exposures to low doses of cobalt nanoparticles induce cell transformation enhanced by
oxidative damage. Nanotoxicology(Early Online). (Supported by Universitat Autònoma de
Barcelona, UAB PIF, the "Generalitat de Catalunya," and the Spanish Ministry of
Education and Science. Authors affiliated with Universitat Autònoma de Barcelona, Spain;
ISCIII, Spain.)
22.
Apostoli P, Porru S, Alessio L. 1994. Urinary cobalt excretion in short time occupational
exposure to cobalt powders. Sci Total Environ 150(1-3): 129-132. (Support not reported.
Authors affiliated with University of Brescia, Italy.)
23.
Arai F, Yamamura Y, Yoshida M, Kishimoto T. 1994. Blood and urinary levels of metals
(Pb, Cr, Cd, Mn, Sb, Co and Cu) in cloisonne workers. Ind Health 32(2): 67-78. (Support
not reported. Authors affiliated with St. Marianna University School of Medicine, Japan.)
24.
Araya J, Maruyama M, Inoue A, Fujita T, Kawahara J, Sassa K, Hayashi R, Kawagishi Y,
Yamashita N, Sugiyama E, Kobayashi M. 2002. Inhibition of proteasome activity is
involved in cobalt-induced apoptosis of human alveolar macrophages. Am J Physiol Lung
Cell Mol Physiol 283(4): L849-858. (Support not reported. Authors affiliated with Toyama
Medical and Pharmaceutical University, Japan.)
25.
Ardyanto TD, Osaki M, Nagahama Y, Yamaga K, Maeta N, Tamura T, Ito H. 2008. Downregulation of cobalt-induced HIF-1alpha expression correlates with cell proliferation and
apoptosis in human gastric carcinoma cells. Oncol Rep 19(2): 339-343. (Support not
reported. Authors affiliated with Tottori University, Japan; Sebelas Maret University,
Indonesia.)
26.
Arlauskas A, Baker RS, Bonin AM, Tandon RK, Crisp PT, Ellis J. 1985. Mutagenicity of
metal ions in bacteria. Environ Res 36(2): 379-388. (as cited in IARC 2006)
27.
Arslan M, Demir H, Arslan H, Gokalp AS, Demir C. 2011. Trace elements, heavy metals
and other biochemical parameters in malignant glioma patients. Asian Pac J Cancer Prev
12(2): 447-451. (Support not reported. Authors affiliated with Yuzuncu Yil University,
Turkey.)
This draft document should not be construed to represent final NTP determination or policy
135
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
28.
Asmuss M, Mullenders LH, Eker A, Hartwig A. 2000. Differential effects of toxic metal
compounds on the activities of Fpg and XPA, two zinc finger proteins involved in DNA
repair. Carcinogenesis 21(11): 2097-2104. (Supported by the Deutsche
Forschungsgemeinschaft. Authors affiliated with University of Karlsruhe, Germany;
Leiden University, Netherlands; Erasmus University Rotterdam, Netherlands.)
29.
ATSDR. 2004. Toxicological Profile for Cobalt. Atlanta, GA: Agency for Toxic
Substances and Disease Registry. pp. 207-E203.
http://www.atsdr.cdc.gov/ToxProfiles/TP.asp?id=373&tid=64.
30.
Auffan M, Rose J, Wiesner MR, Bottero JY. 2009. Chemical stability of metallic
nanoparticles: a parameter controlling their potential cellular toxicity in vitro. Environ
Pollut 157(4): 1127-1133. (Supported by the French National Program ACIFNS (ANR)
‘‘ECCO’’ supported by INSU (French National Institute for Earth Sciences and
Astronomy) and the US National Science Foundation. Authors affiliated with Duke
University, NC; CNRS/Aix-Marseille Université, France; Europôle de l’Arbois, France.)
31.
Ayala-Fierro F, Firriolo JM, Carter DE. 1999. Disposition, toxicity, and intestinal
absorption of cobaltous chloride in male Fischer 344 rats. J Toxicol Environ Health A
56(8): 571-591. (Supported by the Robert S. Flinn Foundation. Authors affiliated with
University of Arizona, AZ; Smith Kline Beecham, NJ.)
32.
Azuma K, Kawahara A, Hattori S, Taira T, Tsurutani J, Watari K, Shibata T, Murakami Y,
Takamori S, Ono M, Izumi H, Kage M, Yanagawa T, Nakagawa K, Hoshino T, Kuwano
M. 2012. NDRG1/Cap43/Drg-1 may predict tumor angiogenesis and poor outcome in
patients with lung cancer. J Thorac Oncol 7(5): 779-789. (Support not reported. Authors
affiliated with Kurume University, Japan; Kinki University, Japan; Kyushu University,
Japan; University of Occupational and Environmental Health, Japan; Graduate School of
Pharmaceutical Sciences, Japan.)
33.
Bailey MR, Kreyling WG, Andre S, Batchelor A, Collier CG, Drosselmeyer E, Ferron GA,
Foster P, Haider B, Hodgson A, Masse R, Metivier H, Morgan A, Müller H-L, Patrick G,
Pearman I, Pickering S, Ramsden D, Stirling C, Talbot RJ. 1989. An interspecies
comparison of the lung clearance of inhaled monodisperse cobalt oxide particles - Part I:
Objectives and summary of results. J Aerosol Sci 20(2): 169-188. (Supported by the
Commission of the European Communities. Authors affiliated with National Radiological
Protection Board, UK; Institut für Strahlenschutz, UK; Institut de Protection et de Sûreté
Nucleaire, France; Medical Research Council Radiobiology Unit, UK; Institut für Genetik
und für Toxikologie von Spaltstoffen, Germany; Atomic Energy Establishment Winfrith,
UK; Harwell Laboratory, UK; European Institute for Transuranium Elements, Germany.)
34.
Bal W, Sokołowska M, Kurowska E, Faller P. 2013. Binding of transition metal ions to
albumin: Sites, affinities and rates. Biochimica Et Biophysica Acta-General Subjects
1830(12): 5444-5455. (Support not reported. Authors affiliated with Polish Academy of
Sciences, Poland; CNRS, France; Université de Toulouse, France.)
35.
Baldwin EL, Byl JA, Osheroff N. 2004. Cobalt enhances DNA cleavage mediated by
human topoisomerase II alpha in vitro and in cultured cells. Biochemistry 43(3): 728-735.
136
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
(Supported by NIH. Authors affiliated with Vanderbilt UniVersity School of Medicine,
TN.)
36.
Banza CL, Nawrot TS, Haufroid V, Decree S, De Putter T, Smolders E, Kabyla BI, Luboya
ON, Ilunga AN, Mutombo AM, Nemery B. 2009. High human exposure to cobalt and other
metals in Katanga, a mining area of the Democratic Republic of Congo. Environ Res
109(6): 745-752. (Support not reported. Authors affiliated with Université de Lubumbashi,
Democratic Republic of Congo; Katholieke Universiteit Leuven, Belgium; Université
catholique de Louvain, Belgium; Royal Museum for Central Africa, Belgium; Facultés
Universitaires Notre-Dame de la Paix, Belgium.)
37.
Barbeau A, Roy M, Paris S. 1984. Hair trace elements in Friedreich's disease. Can J Neurol
Sci 11(4 Suppl): 620-622. (Supported by "L' Association Canadienne pour I' Ataxie de
Friedreich." Authors affiliated with Clinical Research Institute of Montreal, Canada.)
38.
Barrera G. 2012. Oxidative stress and lipid peroxidation products in cancer progression and
therapy. ISRN Oncol 2012: 137289. (Support not reported. Author affiliated with
University of Turin, Italy.)
39.
Basu N, Abare M, Buchanan S, Cryderman D, Nam DH, Sirkin S, Schmitt S, Hu H. 2010.
A combined ecological and epidemiologic investigation of metal exposures amongst
Indigenous peoples near the Marlin Mine in Western Guatemala. Sci Total Environ 409(1):
70-77. (Supported by the Due Process of Law Foundation, the University of Michigan
School of Public Health, Physicians for Human Rights and the Independent International
Panel. Authors affiliated with University of Michigan, MI; University of Illinois at
Chicago, IL; Physicians for Human Rights, MA.)
40.
Bax M. 1981. Lead and impaired abilities. Develop Med Child Neurol 23: 565-566.
(Support and author affiliations not reported.)
41.
Behl M, Stout MD, Herbert RA, Dill JA, Baker GL, Hayden BK, Roycroft JR, Bucher JR,
Hooth MJ. 2015. Comparative toxicity and carcinogenicity of soluble and insoluble cobalt
compounds. Toxicology 333: 195-205. (Support not reported. Authors affiliated with
NIEHS, NC; Battelle Memorial Institute, OH.)
42.
Beleznay E, Osvay M. 1994. Long-term clearance of accidentally inhaled 60Co aerosols in
humans. Health Phys 66(4): 392-399. (Support not reported. Authors affiliated with
Hungarian Academy of Sciences, Hungary.)
43.
Bencko V, Wagner V, Wagnerová M, Zavázal V. 1986. Human exposure to nickel and
cobalt: biological monitoring and immunobiochemical response. Environ Res 40(2): 399410. (Support not reported. Authors affiliated with Charles University, Czechoslovakia;
Regional Hygiene Station of the Central-Bohemian Region, Czechoslovakia; Regional
Institute of National Health, Czechoslovakia.)
44.
Benderli Cihan Y, Öztürk Yildirim S. 2011. A discriminant analysis of trace elements in
scalp hair of healthy controls and stage-IIIB non-small cell lung cancer (NSCLC) patients.
Biol Trace Elem Res 144(1-3): 272-294. (Support not reported. Authors affiliated with
This draft document should not be construed to represent final NTP determination or policy
137
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Kayseri Education and Research Hospital, Turkey; Erciyes University Faculty of Sciences,
Turkey.)
45.
Benderli Cihan Y, Sözen S, Öztürk Yildirim S. 2011. Trace elements and heavy metals in
hair of stage III breast cancer patients. Biol Trace Elem Res 144(1-3): 360-379. (Support
not reported. Authors affiliated with Kayseri Education and Research Hospital, Turkey;
Erciyes University Faculty of Sciences, Turkey.)
46.
Benigni R, Bossa C, Tcheremenskaia O, Battistelli CL, Giuliani A. 2015. The Syrian
hamster embryo cells transformation assay identifies efficiently nongenotoxic carcinogens,
and can contribute to alternative, integrated testing. Mutat Res(In press). (Support not
reported. Authors affiliated with Istituto Superiore di Sanità, Italy.)
47.
Bergomi M, Vinceti M, Nacci G, Pietrini V, Brätter P, Alber D, Ferrari A, Vescovi L,
Guidetti D, Sola P, Malagù S, Aramini C, Vivoli G. 2002. Environmental exposure to trace
elements and risk of amyotrophic lateral sclerosis: a population-based case-control study.
Environ Res 89(2): 116-123. (Supported by the Ministry of the University and of Scientifc
and Technological Research, by the National Research Council, and by the Pietro
Manodori Foundation of Reggio Emilia. Authors affiliated with Università di Modena e
Reggio Emilia, Italy; Università of Parma, Italy; Hahn-Meitner Institut, Germany;
Ospedale Santa Maria Nuova, Italy; Ospedale Bufalini-Marconi, Italy; Ospedale per gli
Infermi, Italy.)
48.
Beyersmann D, Hartwig A. 1992. The genetic toxicology of cobalt. Toxicol Appl
Pharmacol 115(1): 137-145. (Support not reported. Authors affiliated with University of
Bremen, Germany.)
49.
Beyersmann D, Hartwig A. 2008. Carcinogenic metal compounds: recent insight into
molecular and cellular mechanisms. Arch Toxicol 82(8): 493-512. (Support not reported.
Authors affiliated with University of Bremen, Germany; Technical University of Berlin,
Germany.)
50.
Bibi M, Hashmi MZ, Malik RN. 2015. The level and distribution of heavy metals and
changes in oxidative stress indices in humans from Lahore district, Pakistan. Hum Exp
Toxicol. (No funding received. Authors affiliated with Quaid-i-Azam University, Pakistan;
Zhejiang University, China.)
51.
Bourg WJ, Nation JR, Clark DE. 1985. The effects of chronic cobalt exposure on passiveavoidance performance in the adult rat. Bull Psychon Soc 23(6): 527-530. (Supported by
the University Undergraduate Fellows Program at Texas A&M University. Authors
affiliated with Texas A&M University, TX; College Station, TX.)
52.
Bradberry SM, Wilkinson JM, Ferner RE. 2014. Systemic toxicity related to metal hip
prostheses. Clin Toxicol (Phila) 52(8): 837-847. (Support not reported. Authors affiliated
with West Midlands Poisons Unit, UK; University of Sheffield, UK; West Midlands Centre
for Adverse Drug Reactions, UK.)
53.
Bresson C, Darolles C, Carmona A, Gautier C, Sage N, Roudeau S, Ortega R, Ansoborlo E,
Malard V. 2013. Cobalt chloride speciation, mechanisms of cytotoxicity on human
138
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
pulmonary cells, and synergistic toxicity with zinc. Metallomics 5(2): 133-143. (Supported
by the transversal toxicology program from the CEA DSV (France) and Electricité De
France (EDF) company. Authors affiliated with Laboratoire de developpement Analytique
Nucleaire, France; Lab Biochim System Perturb, France; Univ. Bordeaux, France; CNRS,
France; Laboratoire d’Analyse en Soutien aux Exploitants, France; CEA, France.)
54.
Briggs D, Wahlqvist M. 1998. Food Facts Online Edition: Cobalt.
http://apjcn.nhri.org.tw/server/info/books-phds/books/foodfacts/html/data/data-fs.html.
Accessed on 2/23/15.
55.
Broberg K, Engstrom K, Shegufta A. 2015. Gene-environment interactions for metals. In
Handbook on the Toxicology of Metals. 4th ed., Vol. I: General Considerations. Nordberg
GF, Fowler BA, Nordberg M, eds. Waltham, MA: Elsevier. pp. 239-264. (Support and
author affiliations not reported.)
56.
Brock T, Stopford W. 2003. Bioaccessibility of metals in human health risk assessment:
evaluating risk from exposure to cobalt compounds. J Environ Monit 5(4): 71N-76N.
(Support not reported. Authors affiliated with Duke University, NC.)
57.
Bucher JR, Elwell MR, Thompson MB, Chou BJ, Renne R, Ragan HA. 1990. Inhalation
toxicity studies of cobalt sulfate in F344/N rats and B6C3F1 mice. Fundam Appl Toxicol
15(2): 357-372. (Support not reported. Authors affiliated with NIEHS, NC; Pacific
Northwest Laboratories, WA.)
58.
Butt J, Kandel G. 2014. Barrett esophagus: when to endoscope. Clin Endosc 47(1): 40-46.
(Support not reported. Authors affiliated with St. Michael’s Hospital, Canada.)
59.
Caicedo M, Jacobs JJ, Reddy A, Hallab NJ. 2007. Analysis of metal ion-induced DNA
damage, apoptosis, and necrosis in human (Jurkat) T-cells demonstrates Ni2+ and V3+ are
more toxic than other metals: Al3+, Be2+, Co2+, Cr3+, Cu2+, Fe3+, Mo5+, Nb5+, Zr2+. J Biomed
Mater Res Part A: 905-913. (Support not reported. Authors affiliated with Rush University
Medical Center, IL.)
60.
Campbell CA, Peet M, Ward NI. 1988. Vanadium and other trace elements in patients
taking lithium. Biol Psychiatry 24(7): 775-781. (Support not reported. Authors affiliated
with David Rice Hospital, UK; Northern General Hospital, UK; University of Surrey, UK;
Fulbourn Hospital, UK.)
61.
Cancer Research UK. 2014. Oesophageal cancer risk factors. Cancer Research UK.
Updated on 5/21/14. http://www.cancerresearchuk.org/cancerinfo/cancerstats/types/oesophagus/riskfactors/oesophageal-cancer-risk-factors - Overview.
Accessed on 3/30/15.
62.
Carneiro MF, Grotto D, Batista BL, Rhoden CR, Barbosa F, Jr. 2011. Background values
for essential and toxic elements in children's nails and correlation with hair levels. Biol
Trace Elem Res 144(1-3): 339-350. (Supported by the Secretaria de Saúde do Estado do
Rio Grande do Sul, the Conselho Nacional de Desenvolvimento Científico e
Tecnológico— CNP, and the Fundação de Amparo à Pesquisa do Estado de São Paulo.
This draft document should not be construed to represent final NTP determination or policy
139
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Authors affiliated with Universidade de São Paulo, Brazil; Universidade Federal de
Ciências da Saúde de Porto Alegre, Brazil.)
63.
Casto BC, Meyers J, DiPaolo JA. 1979. Enhancement of viral transformation for evaluation
of the carcinogenic or mutagenic potential of inorganic metal salts. Cancer Res 39(1): 193198. (as cited in IARC 2006)
64.
Catelas I, Petit A, Vali H, Fragiskatos C, Meilleur R, Zukor DJ, Antoniou J, Huk OL. 2005.
Quantitative analysis of macrophage apoptosis vs. necrosis induced by cobalt and
chromium ions in vitro. Biomaterials 26(15): 2441-2453. (Supported by the Canadian
Orthopaedic Foundation, the McGill Office of Fellowships and Awards and the
International Federation of University Women. Authors affilliated with McGill University,
Canada; Beckman Coulter Canada, Inc., Canada.)
65.
Cavallo D, Ciervo A, Fresegna AM, Maiello R, Tassone P, Buresti G, Casciardi S, Iavicoli
S, Ursini CL. 2015. Investigation on cobalt-oxide nanoparticles cyto-genotoxicity and
inflammatory response in two types of respiratory cells. J Appl Toxicol. (Supported by the
Italian Ministry of Health. Authors affiliated with Italian Workers’ Compensation
Authority, Italy.)
66.
CDC. 2013. Biomonitoring Summary: cobalt. Centers for Disease Control and Prevention.
http://www.cdc.gov/biomonitoring/Cobalt_BiomonitoringSummary.html. Accessed on
1/31/15.
67.
CDC. 2015. Fourth National Report on Human Exposure to Environmental Chemicals,
Updated Tables, February 2015. Atlanta, GA: Centers for Disease Control and Prevention.
1095 pp.
68.
CDI. 2006. Cobalt Facts. Cobalt Development Institute. http://thecdi.com/cobaltfacts.php.
Accessed on 2/12/15.
69.
Cereda C, Redaelli ML, Canesi M, Carniti A, Bianchi S. 1994. Widia tool grinding: the
importance of primary prevention measures in reducing occupational exposure to cobalt.
Sci Total Environ 150(1-3): 249-251. (Support not reported. Authors affiliated with Health
and Safety Unit, Italy.)
70.
Chadwick JK, Wilson HK, White MA. 1997. An investigation of occupational metal
exposure in thermal spraying processes. Sci Total Environ 199(1-2): 115-124. (Support not
reported. Authors affiliated with Health and Safety Laboratory, UK; Commission of the
European Communities Joint Research Centre, Italy.)
71.
Chattopadhyay S, Dash SK, Tripathy S, Das B, Mandal D, Pramanik P, Roy S. 2015.
Toxicity of cobalt oxide nanoparticles to normal cells; an in vitro and in vivo study. Chem
Biol Interact 226: 58-71. (Support not reported. Authors affiliated with Vidyasagar
University, India; Indian Institute of Technology, India.)
72.
ChemIDplus. 2015. ChemIDplus Lite. National Library of Medicine.
http://chem.sis.nlm.nih.gov/chemidplus/chemidlite.jsp and search by CAS number.
Accessed on 3/24/15.
140
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
73.
Cheng Y, Chen G, Hong L, Zhou L, Hu M, Li B, Huang J, Xia L, Li C. 2013. How does
hypoxia inducible factor-1alpha participate in enhancing the glycolysis activity in cervical
cancer? Ann Diagn Pathol 17(3): 305-311. (Supported by China's Post-doctoral Science
Fund and the Natural Science Fund of Hubei Province. Authors affiliated with Renmin
Hospital of Wuhan University, China; Hospital of Xiantao City, China; Third Affiliated
Hospital of Guangzhou Medical College, China.)
74.
Cheyns K, Banza Lubaba Nkulu C, Ngombe LK, Asosa JN, Haufroid V, De Putter T,
Nawrot T, Kimpanga CM, Numbi OL, Ilunga BK, Nemery B, Smolders E. 2014. Pathways
of human exposure to cobalt in Katanga, a mining area of the D.R. Congo. Sci Total
Environ 490: 313-321. (Support not reported. Authors affiliated with Veterinary and
Agrochemical Research Centre, Belgium; Université de Lubumbashi, People's Republic of
Congo; Université de Kamina, People's Republic of Congo; Université Catholique de
Louvain, Belgium; Royal Museum for Central Africa, Belgium; Department of Public
Health and Primary Care, Belgium; Hasselt University, Belgium; KU Leuven, Belgium.)
75.
Chong R, Ke-zhou C, Zeng-liang Y. 2009. Induction of germline apoptosis by cobalt and
relevant signal transduction pathways in Caenorhabditis elegans. Toxicol Mech Methods
19(9): 541-546. (Support not reported. Authors affiliated with Chinese Academy of
Sciences, China.)
76.
Christensen JM, Mikkelsen S. 1986. Cobalt concentration in whole blood and urine from
pottery plate painters exposed to cobalt paint. In Proceedings of an International
Conference, Heavy Metals in the Environment, Athens, September 1985. Lakkas TD, ed.
Luxembourg: Commission of the European Communities. pp. 86-88. (as cited in IARC
1991)
77.
Christensen JM, Poulsen OM, Thomsen M. 1993a. A short-term cross-over study on oral
administration of soluble and insoluble cobalt compounds: sex differences in biological
levels. Int Arch Occup Environ Health 65(4): 233-240. (Supported by Royal Copenhagen
A/S and the Danish Working Environment Fund. Authors affiliated with National Institute
of Occupational Health, Denmark; Danish Emergency Management Agency, Denmark.)
78.
Christensen JM, Poulsen OM. 1994. A 1982-1992 surveillance programme on Danish
pottery painters. Biological levels and health effects following exposure to soluble or
insoluble cobalt compounds in cobalt blue dyes. Sci Total Environ 150(1-3): 95-104.
(Support not reported. Authors affiliated with Danish National Institute of Occupational
Health, Denmark.)
79.
Christensen JM. 1995. Human exposure to toxic metals: factors influencing interpretation
of biomonitoring results. Sci Total Environ 166: 89-135. (Support not reported. Authors
affiliated with National Institute of Occupational Health, Denmark.)
80.
Clyne N, Lins L-E, Pehrsson SK, Lundberg A, Werner J. 1988. Distribution of cobalt in
myocardium, skeletal muscle and serum in exposed and unexposed rats. Trace Elem Med
5(2): 52-54. (Supported by the Magnus Bergvall's and Tore Nilsson's foundations. Authors
affiliated with Karolinska Hospital, Sweden; Swedish Environmental Research Institute,
Sweden.)
This draft document should not be construed to represent final NTP determination or policy
141
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
81.
Coleman RF, Herrington J, Scales JT. 1973. Concentration of wear products in hair, blood,
and urine after total hip replacement. Br Med J 1(5852): 527-529. (as cited in Schaffer et al
1999)
82.
Collecchi P, Esposito M, Brera S, Mora E, Mazzucotelli A, Oddone M. 1986. The
distribution of arsenic and cobalt in patients with laryngeal carcinoma. J Appl Toxicol 6(4):
287-289. (Support not reported. Authors affiliated with Istituto Nazionale per la Ricerca sul
Cancro, Italy; Clinica Otorinolaringoiatrica ‘B dell’ Universita, Italy; Dipartimento di
Chimica Generale dell’ Universita di Pavia, Italy.)
83.
Collier CG, Bailey MR, Hodgson A. 1989. An interspecies comparison of the lung
clearance of inhaled monodisperse cobalt oxide particles - Part V: Lung clearance of
inhaled cobalt oxide particles in hamster, rats and guinea pigs. J Aerosol Sci 20(2): 233247. (Supported by the CEC. Authors affiliated with National Radiological Protection
Board (NRPB), UK.)
84.
Collier CG, Hodgson A, Gray A, Moody JC, Ball A. 1991. The lung clearance kinetics of
57
Co3O4 in rats of various ages. J Aerosol Sci 22(4): 537-549. (Supported by the CEC.
Authors affiliated with National Radiological Protection Board (NRPB), UK.)
85.
Colognato R, Bonelli A, Ponti J, Farina M, Bergamaschi E, Sabbioni E, Migliore L. 2008.
Comparative genotoxicity of cobalt nanoparticles and ions on human peripheral leukocytes
in vitro. Mutagenesis 23(5): 377-382. (Supported by the Italian Ministry of University and
Scientific Research. Authors affiliated with University of Pisa, Italy; European
Commission, Italy; University of Parma, Italy.)
86.
Cook MB, Kamangar F, Whiteman DC, Freedman ND, Gammon MD, Bernstein L, Brown
LM, Risch HA, Ye W, Sharp L, Pandeya N, Webb PM, Wu AH, Ward MH, Giffen C,
Casson AG, Abnet CC, Murray LJ, Corley DA, Nyren O, Vaughan TL, Chow WH. 2010.
Cigarette smoking and adenocarcinomas of the esophagus and esophagogastric junction: a
pooled analysis from the international BEACON consortium. J Natl Cancer Inst 102(17):
1344-1353. (Supported by NIH, the Chief Scientist Office (Scotland), the Locally
Organised Research Scheme, the Special Trustees of the Nottingham University Hospitals,
the Medical Research Council, the California Tobacco-Related Research Program, NCI, the
Nova Scotia Health Research Foundation, the Northern Ireland Research and Development
Office, the Health Research Board, Ireland, the Queensland Cancer Fund, and the National
Health and Medical Research Council of Australia. Authors affiliated with NIH, MD;
Morgan State University, MD; Queensland Institute of Medical Research, Australia;
University of North Carolina School of Public Health, NC; Beckman Research Institute and
City of Hope Comprehensive Cancer Center, CA; RTI International, MD; Yale University
School of Medicine, CT; Karolinska Institutet, Sweden; National Cancer Registry, Ireland;
USC Norris Comprehensive Cancer Center, CA; Information Management Services, MD;
University of Saskatchewan, Canada; Queen’s University, Ireland; Kaiser Permanente, CA;
Fred Hutchinson Cancer Research Center, WA.)
87.
Coombs M. 1996. Biological monitoring of cobalt oxide workers. Int Arch Occup Environ
Health 68(6): 511-512. (Support not reported. Authors affiliated with Sentrachem Ltd.,
South Africa.)
142
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
88.
Corley DA, Levin TR, Habel LA, Weiss NS, Buffler PA. 2002. Surveillance and survival
in Barrett's adenocarcinomas: a population-based study. Gastroenterology 122(3): 633-640.
(Supported by the American Digestive Health Foundation/TAP Outcomes Research
Award, and the National Institutes of Health Mentored Clinical Scientist Development
Award. Authors affiliated with University of California - Berkely and San Francisco, CA;
Northern California Kaiser Division of Research, CA; University of Washington, WA.)
89.
Costa M, Heck JD, Robison SH. 1982. Selective phagocytosis of crystalline metal sulfide
particles and DNA strand breaks as a mechanism for the induction of cellular
transformation. Cancer Res 42(7): 2757-2763. (Supported by the Office of Research and
Development, Environmental Protection Agency, NIEHS and the the United States
Department of Energy. Authors affiliated with University of Texas, TX. )
90.
Coyle YM, Hynan LS, Euhus DM, Minhajuddin AT. 2005. An ecological study of the
association of environmental chemicals on breast cancer incidence in Texas. Breast Cancer
Res Treat 92(2): 107-114. (Supported by the Natalie Ornish Fund, the Clay Weed
Memorial Trust Fund, and CREW Dallas. Authors affiliated with University of Texas
Southwestern Medical Center, TX.)
91.
Coyle YM, Minahjuddin AT, Hynan LS, Minna JD. 2006. An ecological study of the
association of metal air pollutants with lung cancer incidence in Texas. Journal of Thoracic
Oncology 1(7): 654-661. (Supported by the SPORE in Lung Cancer and the Clay Weed
Memorial Trust Fund. Authors affiliated with University of Texas Southwestern Medical
Center, TX.)
92.
Davidson T, Ke Q, Costa M. 2015. Selected molecular mechanisms of metal toxicity and
carcinogenicity. In Handbook on the Toxicology of Metals. 4th ed., Vol. I: General
Considerations. Nordberg GF, Fowler BA, Nordberg M, eds. Waltham, MA: Elsevier. pp.
173-196.
93.
Davies AP, Sood A, Lewis AC, Newson R, Learmonth ID, Case CP. 2005. Metal-specific
differences in levels of DNA damage caused by synovial fluid recovered at revision
arthroplasty. J Bone Joint Surg Br 87(10): 1439-1444. (Support not reported. Authors
affiliated with Southmead Hospital, UK; King’s College Hospital, UK.)
94.
Davis JR, ed. 2000. Nickel, Cobalt, and Their Alloys. Materials Park, OH: ASM
International. pp. 345-406.
95.
De Boeck M, Lison D, Kirsch-Volders M. 1998. Evaluation of the in vitro direct and
indirect genotoxic effects of cobalt compounds using the alkaline comet assay. Influence of
interdonor and interexperimental variability. Carcinogenesis 19(11): 2021-2029.
(Supported by by the Belgian Federal Offices for Scientific, Technical and Cultural Affairs
(OSTC) of the Prime Minister’s Services. Authors affiliated with Vrije Universiteit
Brussel, Belgium; Université Catholique de Louvain, Belgium.)
96.
De Boeck M, Lardau S, Buchet JP, Kirsch-Volders M, Lison D. 2000. Absence of
significant genotoxicity in lymphocytes and urine from workers exposed to moderate levels
of cobalt-containing dust: a cross-sectional study. Environ Mol Mutagen 36(2): 151-160.
(Supported by the Belgian Federal Offices for Scientific, Technical and Cultural Affairs
This draft document should not be construed to represent final NTP determination or policy
143
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
and Cobalt Development Institute. Authors affiliated with Vrije Universiteit Brussel,
Belgium; Université catholique de Louvain, Belgium.)
97.
De Boeck M, Kirsch-Volders M, Lison D. 2003a. Cobalt and antimony: genotoxicity and
carcinogenicity. Mutat Res 533(1-2): 135-152. (Support not reported. Authors affiliated
with Vrije Universiteit Brussel, Belgium; Université Catholique de Louvain, Belgium.)
98.
De Boeck M, Lombaert N, De Backer S, Finsy R, Lison D, Kirsch-Volders M. 2003b. In
vitro genotoxic effects of different combinations of cobalt and metallic carbide particles.
Mutagenesis 18(2): 177-186. (Supported by the the Belgian Federal Offices for Scientific,
Technical and Cultural Affairs of the Prime Minister’s Service. Authors affiliated with
Vrije Universiteit Brussel, Belgium; Université Catholique de Louvain, Belgium.)
99.
De Olivera JV, Boufleur LA, Dos Santos CE, Dias JF, Squeff CH, Silva GR, Ianistcki M,
Benvegnu VC, Da Silva J. 2012. Occupational genotoxicity among copper smelters.
Toxicol Ind Health 28(9): 789-795. (Support not received. Authors affiliated with
Universidade Luterana do Brasil, Brazil; Universidade Federal do Rio Grande do Sul,
Brazil.)
100. De Palma G, Goldoni M, Catalani S, Carbognani P, Poli D, Mozzoni P, Acampa O,
Internullo E, Rusca M, Apostoli P. 2008. Metallic elements in pulmonary biopsies from
lung cancer and control subjects. Acta Biomed 79 Suppl 1: 43-51. (Support not reported.
Authors affiliated with University of Brescia, Italy; University of Parma, Italy; University
Hospital of Parma, Italy.)
101. Delahant AB. 1955. An experimental study of the effects of rare metals on animal lungs.
AMA Arch Ind Health 12(2): 116-120. (Support not reported. Author affiliated with
Saranac Laboratory, NY.)
102. Della Torre F, Cassani M, Segale M, Scarpazza G, Pietra R, Sabbioni E. 1990. Trace metal
lung diseases: a new fatal case of hard metal pneumoconiosis. Respiration 57(4): 248-253.
(Support not reported. Authors affiliated with INRCA Casatenovo, Italy; Commission of
the European Communities, Italy.)
103. Demir C, Demir H, Esen R, Sehitogullari A, Atmaca M, Alay M. 2011. Altered serum
levels of elements in acute leukemia cases in Turkey. Asian Pac J Cancer Prev 12(12):
3471-3474. (Suport not reported. Authors affiliated with Yuzuncu Yil University, Turkey;
Van Training and Research Hospital, Turkey.)
104. Deng X, Tüysüz H. 2014. Cobalt-oxide-based materials as water oxidation catalyst: Recent
progress and challenges. ACS Catal 4: 3701-3714. (Support not reported. Authors affiliated
with Max-Planck Institut für Kohlenforschung, Germany.)
105. Desai TK, Krishnan K, Samala N, Singh J, Cluley J, Perla S, Howden CW. 2012. The
incidence of oesophageal adenocarcinoma in non-dysplastic Barrett's oesophagus: a metaanalysis. Gut 61(7): 970-976. (Support not reported. Authors affiliated with William
Beaumont Hospital, MI; Northwestern University Feinberg School of Medicine, IL;
Oakland University, MI. Dr. Howard is a consultant for Takeda, Otsuka, Boehringer-
144
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Ingelheim, Novartis, Eisai, and XenoPort; and he has received speaking fees from Takeda,
Otsuka, and GlaxoSmithKline)
106. Devlin JJ, Pomerleau AC, Brent J, Morgan BW, Deitchman S, Schwartz M. 2013. Clinical
features, testing, and management of patients with suspected prosthetic hip-associated
cobalt toxicity: a systematic review of cases. J Med Toxicol 9(4): 405-415. (Support not
reported. Authors affiliated with Georgia Poison Center, GA; Emory University School of
Medicine, GA; Centers for Disease Control and Prevention, GA; University of Colorado
School of Medicine, CO; DePuy Companies.)
107. Dick CA, Brown DM, Donaldson K, Stone V. 2003. The role of free radicals in the toxic
and inflammatory effects of four different ultrafine particle types. Inhal Toxicol 15(1): 3952. (Support not reported. Authors affiliated with Napier University, UK; Scottish
Biomedical, UK.)
108. Dongarrà G, Lombardo M, Tamburo E, Varrica D, Cibella F, Cuttitta G. 2011.
Concentration and reference interval of trace elements in human hair from students living
in Palermo, Sicily (Italy). Environ Toxicol Pharmacol 32(1): 27-34. (Supported by Miur
and CNR - IBIM: Institute of Biomedicine and MolecularImmunology “A. Monroy”.
Authors affiliated with Università di Palermo, Italy; Consiglio Nazionale delle Ricerche,
Italy.)
109. Doran A, Law FC, Allen MJ, Rushton N. 1998. Neoplastic transformation of cells by
soluble but not particulate forms of metals used in orthopaedic implants. Biomaterials
19(7-9): 751-759. (as cited in IARC 2006)
110. Dunstan E, Sanghrajka AP, Tilley S, Unwin P, Blunn G, Cannon SR, Briggs TW. 2005.
Metal ion levels after metal-on-metal proximal femoral replacements: a 30-year follow-up.
J Bone Joint Surg Br 87(5): 628-631. (Support not reported. Authors affiliated with Royal
National Orthopaedic Hospital, UK.)
111. ECHA. 2009. Practical Guide 6: How to Report Read-Across and Categories. ECHA-10B-11-EN. Helsinki, Finland: European Chemicals Agency. 34 pp.
112. Edel J, Pozzi G, Sabbioni E, Pietra R, Devos S. 1994. Metabolic and toxicological studies
on cobalt. Sci Total Environ 150(1-3): 233-244. (Support not reported. Authors affiliated
with European Commission, Italy.)
113. Egilsson V, Evans IH, Wilkie D. 1979. Toxic and mutagenic effects of carcinogens on the
mitochondria of Saccharomyces cerevisiae. Mol Gen Genet 174(1): 39-46. (as cited in
IARC 2006)
114. Elenge MM, Aubry JC, Jacob L, De Brouwer C. 2011. Heavy metal in hair samples of 109
non-industrial (miners) population in Katanga. Sante 21(1): 41-46. (No funds received.
Authors affiliated with Universite Libre de Bruxelles, Belgique.)
115. EPA. 2012. EPA Chemical Data Reporting. U.S. Environmental Protection Agency.
http://java.epa.gov/oppt_chemical_search/ and search by CAS number. Accessed on
3/24/15.
This draft document should not be construed to represent final NTP determination or policy
145
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
116. EPA. 2015. EPA Substance Registry Services. U.S. Environmental Protection Agency.
http://iaspub.epa.gov/sor_internet/registry/substreg/searchandretrieve/substancesearch/sear
ch.do and search on CAS number. Accessed on 3/24/15.
117. Erturk FA, Ay H, Nardemir G, Agar G. 2013. Molecular determination of genotoxic effects
of cobalt and nickel on maize (Zea mays L.) by RAPD and protein analyses. Toxicol Ind
Health 29(7): 662-671. (Support not reported. Authors affiliated with Atatürk University,
Turkey.)
118. Fairhall LT, Keenan RG, Brinton HP. 1949. Cobalt and the dust environment of the
cemented tungsten carbide industry. Public Health Rep 64(15): 485-490. (Support not
reported. Authors affiliated with Public Health Service.)
119. Farah SB. 1983. The in vivo effect of cobalt chloride on chromosomes. Rev Brasil Genet 6:
433-442. (as cited in IARC 2006)
120. Farquharson JP. 2015. Formation Metals Inc. - President's Letter to Shareholders.
Formation Metals, Inc. Updated on 2/23/15.
http://www.formationmetals.com/s/CobaltNews.asp?ReportID=697830. Accessed on
3/6/15.
121. Ferdenzi P, Giaroli C, Mori P, Pedroni C, Piccinini R, Ricci R, Sala O, Veronesi C, Mineo
F. 1994. Cobalt powdersintering industry (stone cutting diamond wheels): a study of
environmental-biological monitoring, workplace improvement and health surveillance. Sci
Total Environ 150(1-3): 245-248. (Support not reported. Authors affiliated with SMPIL
USL 9, Italy; USL 12 Scandiano, Italy; Industrial Physician, Italy; PMP USL 9, Italy.)
122. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JW, Comber H,
Forman D, Bray F. 2013. Cancer incidence and mortality patterns in Europe: estimates for
40 countries in 2012. Eur J Cancer 49(6): 1374-1403. (Supported by the ERA-NET project
EUROCOURSE funded within the Seventh Framework Programme of the European
Commission. Authors affiliated with IARC, France; CPO – Centre for Epidemiology and
Prevention in Oncology in Piedmont, Italy; Erasmus MC, Netherlands; Comprehensive
Cancer Centre South (IKZ), Netherlands; National Cancer Registry, Ireland.)
123. Ferri F, Candela S, Bedogni L, Piccinini R, Sala O. 1994. Exposure to cobalt in the welding
process with stellite. Sci Total Environ 150(1-3): 145-147. (Support not reported. Authors
affiliated with Servizio di Medicina Preventiva ed Igiene del Lavoro, Italy; Presidio
Multizonale di Prevenzione, Italy.)
124. Figgitt M, Newson R, Leslie IJ, Fisher J, Ingham E, Case CP. 2010. The genotoxicity of
physiological concentrations of chromium (Cr(III) and Cr(VI)) and cobalt (Co(II)): an in
vitro study. Mutat Res 688(1-2): 53-61. (Supported by the Arthritis and Rheumatism
Council. Authors affiliated with Southmead Hospital, UK; Imperial College London, UK;
University of Leeds, UK.)
125. Finley BL, Monnot AD, Gaffney SH, Paustenbach DJ. 2012. Dose-response relationships
for blood cobalt concentrations and health effects: a review of the literature and application
146
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
of a biokinetic model. J Toxicol Environ Health B Crit Rev 15(8): 493-523. (Supported by
DePuy Orthopedics, Inc. Authors affiliated with ChemRisk, LLC, CA.)
126. Finley BL, Unice KM, Kerger BD, Otani JM, Paustenbach DJ, Galbraith DA, Tvermoes
BE. 2013. 31-day study of cobalt(II) chloride ingestion in humans: pharmacokinetics and
clinical effects. J Toxicol Environ Health A 76(21): 1210-1224. (Supported by DePuy
Orthopedics, Inc. Authors affiliated with Cardno ChemRisk, CA, PA and CO.)
127. Finogenova MA. 1973. [The influence of cobalt on induced cutaneous carcinogenesis].
Biull Eksp Biol Med 74(2): 73-75. (Support not reported. Authors affiliated with Nutrition
Institute of the USSR Academy of Medicine, Russia.)
128. Fleckman P. 1985. Anatomy and physiology of the nail. Dermatol Clin 3(3): 373-381.
(Support not reported. Author affiliated with University of Washington, WA.)
129. Foster PP, Pearman I, Ramsden D. 1989. An interspecies comparison of the lung clearance
of inhaled monodisperse cobalt oxide particles - Part II: Lung clearance of inhaled cobalt
oxide particles in man. J Aerosol Sci 20(2): 189-204. (Supported by the UKAEA Core
Programme on Radiological Protection Research and the CEC. Authors affiliated with
Atomic Energy Establishment, UK.)
130. Frank E. 2012. Metal-on-Metal Hip Systems. U.S. Food and Drug Administration.
http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Medic
alDevices/MedicalDevicesAdvisoryCommittee/OrthopaedicandRehabilitationDevicesPanel
/UCM310217.pdf target=.
131. Freedman ND, Murray LJ, Kamangar F, Abnet CC, Cook MB, Nyren O, Ye W, Wu AH,
Bernstein L, Brown LM, Ward MH, Pandeya N, Green AC, Casson AG, Giffen C, Risch
HA, Gammon MD, Chow WH, Vaughan TL, Corley DA, Whiteman DC. 2011. Alcohol
intake and risk of oesophageal adenocarcinoma: a pooled analysis from the BEACON
Consortium. Gut 60(8): 1029-1037. (Supported by NIH, NCI, the Nova Scotia Health
Research Foundation, the Northern Ireland Research & Development Office, the Health
Research Board, Ireland, the California Tobacco Related Research Program, the
Queensland Cancer Fund, and the National Health and Medical Research Council
(NHMRC) of Australia. Authors affiliated with NIH, MD; Queen’s University, Ireland;
Morgan State University, MD; Karolinska Institutet, Sweden; University of Southern
California/Norris Comprehensive Cancer Center, CA; Beckman Research Institute, CA;
RTI International, MD; Queensland Institute of Medical Research, Australia; University of
Saskatchewan, Canada; Information Management Services, MD; Yale University School of
Medicine, CT; University of North Carolina School of Public Health, NC; Fred Hutchinson
Cancer Research Center, WA; Kaiser Permanente, CA.)
132. Fu OY, Hou MF, Yang SF, Huang SC, Lee WY. 2009. Cobalt chloride-induced hypoxia
modulates the invasive potential and matrix metalloproteinases of primary and metastatic
breast cancer cells. Anticancer Res 29(8): 3131-3138. (Supported by the National Science
Council and the and Chi Mei Medical Center. Authors affiliated with Kaohsiung Medical
University, Taiwan; Chi Mei Medical Center, Taiwan; Taipei Medical University, Taiwan;
Southern Taiwan University, Taiwan.)
This draft document should not be construed to represent final NTP determination or policy
147
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
133. Fukunaga M, Kurachi Y, Mizuguchi Y. 1982. Action of some metal ions on yeast
chromosomes. Chem Pharm Bull (Tokyo) 30(8): 3017-3019. (as cited in IARC 2006)
134. Galán-Cobo A, Sánchez-Silva R, Serna A, Abreu-Rodríguez I, Muñoz-Cabello AM,
Echevarría M. 2013. Cellular overexpression of Aquaporins slows down the natural HIF-2α
degradation during prolonged hypoxia. Gene 522(1): 18-26. (Supported by the “Instituto de
Salud Carlos III," “La Junta de Andalucía”, Consejería de Salud, Consejería de Innovación
Ciencia y Empresa, "La Junta de Andalucía, Consejería de Innovación Ciencia y Empresa”,
Spain and the Spanish Ministry of Science and Innovation. Authors affiliated with Hospital
Universitario Virgen del Rocío/CSIC/Universidad de Sevilla, Spain; Centro de
Investigación Biomédica en Red sobre Enfermedades Respiratorias (CIBERES), Spain.)
135. Galanis A, Pappa A, Giannakakis A, Lanitis E, Dangaj D, Sandaltzopoulos R. 2008.
Reactive oxygen species and HIF-1 signalling in cancer. Cancer Lett 266(1): 12-20.
(Supported by a Marie Curie International Reintegration Grant within the 6th European
Community Framework Program and the Hellenic Ministry of Education. Authors
affiliated with Democritus University of Thrace, Greece.)
136. Galanis A, Karapetsas A, Sandaltzopoulos R. 2009. Metal-induced carcinogenesis,
oxidative stress and hypoxia signalling. Mutation Research - Genetic Toxicology and
Environmental Mutagenesis 674(1-2): 31-35. (Support not reported. Authors affiliated with
Democritus University of Thrace, Greece.)
137. Gao S, Zhou J, Zhao Y, Toselli P, Li W. 2013. Hypoxia-response element (HRE)-directed
transcriptional regulation of the rat lysyl oxidase gene in response to cobalt and cadmium.
Toxicol Sci 132(2): 379-389. (Supported by NIH. Authors affiliated with Boston University
School of Medicine, MA.)
138. Garland M, Morris JS, Rosner BA, Stampfer MJ, Spate VL, Baskett CJ, Willett WC,
Hunter DJ. 1993. Toenail trace element levels as biomarkers: reproducibility over a 6-year
period. Cancer Epidemiol Biomarkers Prev 2(5): 493-497. (Supported by NIH and NIEHS.
Authors affiliated with Harvard School of Public Health, MA; Harvard Medical School and
Brigham and Women’s Hospital, MA.)
139. Garrick MD, Dolan KG, Horbinski C, Ghio AJ, Higgins D, Porubcin M, Moore EG,
Hainsworth LN, Umbreit JN, Conrad ME, Feng L, Lis A, Roth JA, Singleton S, Garrick
LM. 2003. DMT1: a mammalian transporter for multiple metals. Biometals 16(1): 41-54.
(Supported by the NIH, USDA, NSF and EPA. Authors affiiated with SUNY at Buffalo,
NY; Environmental Protection Agency, NC; University of South Alabama, AL.)
140. Gerhardsson L, Wester PO, Nordberg GF, Brune D. 1984. Chromium, cobalt and
lanthanum in lung, liver and kidney tissue from deceased smelter workers. Sci Total
Environ 37(2-3): 233-246. (Supported by the Swedish Work Environment Fund. Authors
affiliated with Umeå University, Sweden; Scandinavian Institute of Dental Materials,
Norway.)
141. Gerhardsson L, Brune D, Nordberg IG, Wester PO. 1985. Protective effect of selenium on
lung cancer in smelter workers. Br J Ind Med 42(9): 617-626. (Supported by the Swedish
148
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Work Environment Fund. Authors affiliated with University of Umeå, Sweden; University
Hospital, Sweden; Scandanavian Institute of Dental Materials, Norway.)
142. Gerhardsson L, Brune D, Lundstrom NG, Nordberg G, Wester PO. 1993. Biological
specimen bank for smelter workers. Sci Total Environ 140: 157-173. (Supported by the
Swedish Work Environment Fund. Authors affiliated with University of Umeå, Sweden;
Lund University Hospital, Sweden; D.B. C., Dag Brune Consulting, Norway; University
Hospital of Umeå, Sweden.)
143. Gerhardsson L, Nordberg GF. 1993. Lung cancer in smelter workers--interactions of metals
as indicated by tissue levels. Scand J Work Environ Health 19 Suppl 1: 90-94. (Support not
reported. Authors affiliated with University of Umeå, Sweden; University Hospital,
Sweden.)
144. Gilbert EW, Luna RA, Harrison VL, Hunter JG. 2011. Barrett's esophagus: a review of the
literature. J Gastrointest Surg 15(5): 708-718. (Support not reported. Authors affiliated
with Oregon Health Sciences University, OR.)
145. Gilman JP. 1962. Metal carcinogenesis. II. A study on the carcinogenic activity of cobalt,
copper, iron, and nickel compounds. Cancer Res 22: 158-162. (Supported by the National
Cancer Institute of Canada. Author affiliated with Ontario Veterinary College, Canada.)
146. Gilman JP, Ruckerbauer GM. 1962. Metal carcinogenesis. I. Observations on the
carcinogenicity of a refinery dust, cobalt oxide, and colloidal thorium dioxide. Cancer Res
22: 152-157. (Supported by the National Cancer Institute of Canada. Authors affiliated
with Ontario Veterinary College, Canada.)
147. González-Muñoz MJ, Sánchez-Muniz FJ, Ródenas S, Sevillano MI, Larrea Marín MT,
Bastida S. 2010. Differences in metal and metalloid content in the hair of normo- and
hypertensive postmenopausal women. Hypertens Res 33(3): 219-224. (Supported by
projects PR248/01-10161 and AGL Consolider- Ingenio 2010 project reference CSD200700016. Authors affiliated with Universidad de Alcalá, Spain; Universidad Complutense,
Spain; Universidad Complutense de Madrid, Spain; CSIC, Spain.)
148. Gori C, Zucconi L. 1957. [L’azione citologica indotta da un gruppo di composti inorganici
su Allium cepa]. Caryologia 10: 29-45. (as cited in IARC 2006)
149. Granchi D, Cenni E, Ciapetti G, Savarino L, Stea S, Gamberini S, Gori A, Pizzoferrato A.
1998. Cell death induced by metal ions: necrosis or apoptosis? J Mater Sci Mater Med
9(1): 31-37. (Supported by the Istituti Ortopedici Rizzoli, 'Ricerca Corrente' and Finalized
Project 'Citocompatibilità Quantitativa nella Valutazione dei Materiali Protesici in
Chirurgia Ortopedica'. Authors affiliated with Istituti Ortopedici Rizzoli, Italy.)
150. Green SE, Luczak MW, Morse JL, DeLoughery Z, Zhitkovich A. 2013. Uptake, p53
pathway activation, and cytotoxic responses for Co(II) and Ni(II) in human lung cells:
implications for carcinogenicity. Toxicol Sci 136(2): 467-477. (Supported by NIEHS.
Authors affiliated with Brown University, RI.)
This draft document should not be construed to represent final NTP determination or policy
149
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
151. Gregus Z, Klaassen CD. 1986. Disposition of metals in rats: a comparative study of fecal,
urinary, and biliary excretion and tissue distribution of eighteen metals. Toxicol Appl
Pharmacol 85(1): 24-38. (Supported by USPHS. Authors affiliated with University of
Kansas, KS.)
152. Greim H, Hartwig A, Reuter U, Richter-Reichhelm HB, Thielmann HW. 2009. Chemically
induced pheochromocytomas in rats: Mechanisms and relevance for human risk
assessment. Critical Reviews in Toxicology 39(8): 695-718. (Supported by the Deutsche
Forschungsgemeinschaft. Authors affiliated with Technical University of Munich,
Germany; Technical University of Berlin, Germany; Bundesinstitut für Risikobewertung,
Germany; German Cancer Research Center, Germany.)
153. Grimsrud TK, Berge SR, Haldorsen T, Andersen A. 2002. Exposure to different forms of
nickel and risk of lung cancer. Am J Epidemiol 156(12): 1123-1132. (Supported by the
Norwegian Cancer Society, the Confederation of Norwegian Business and Industry
Working Environment Fund, Falconbridge Nikkelverk A/S, and the Cancer Registry of
Norway. Authors affiliated with Cancer Registry of Norway, Norway; Falconbridge
Nikkelverk A/S, Norway.)
154. Grimsrud TK, Berge SR, Martinsen JI, Andersen A. 2003. Lung cancer incidence among
Norwegian nickel-refinery workers 1953-2000. J Environ Monit 5(2): 190-197. (Supported
by the Confederation of Norwegian Business and Industry Working Environment Fund.
Authors affiliated with Cancer Registry of Norway, Norway; Falconbridge Nikkelverk A/S,
Norway.)
155. Grimsrud TK, Berge SR, Haldorsen T, Andersen A. 2005. Can lung cancer risk among
nickel refinery workers be explained by occupational exposures other than nickel?
Epidemiology 16(2): 146-154. (Supported by the Norwegian Cancer Society, the
Confederation of Norwegian Business and Industry (CNBI) Working Environment Fund,
Falconbridge Nikkelverk A/S, and the Cancer Registry of Norway. Authors affiliated with
Cancer Registry of Norway, Norway; Falconbridge Nikkelverk A/S, Norway.)
156. Gunn SA, Gould TC, Anderson WA. 1967. Specific response of mesenchymal tissue to
cancerigenesis by cadmium. Arch Pathol 83(6): 493-499. (Supported by NCI. Authors
affiliated with University of Miami School of Medicine, FL.)
157. Guskov A, Eshaghi S. 2012. The mechanisms of Mg2+ and Co2+ transport by the CorA
family of divalent cation transporters. Curr Top Membr 69: 393-414. (Supported by the
National Research Foundation and Biomedical Research Council of Singapore. Authors
affiliated with Nanyang Technological University, Singapore.)
158. Guyton KZ, Kyle AD, Aubrecht J, Cogliano VJ, Eastmond DA, Jackson M, Keshava N,
Sandy MS, Sonawane B, Zhang L, Waters MD, Smith MT. 2009. Improving prediction of
chemical carcinogenicity by considering multiple mechanisms and applying toxicogenomic
approaches. Mutat Res 681(2-3): 230-240. (Supported by the U.S. EPA, ILS, Inc, Pfizer,
Inc., and NIEHS. Authors affiliated with U.S. EPA, Washington, D.C.; University of
California - Berkeley, CA; Pfizer Global Research and Development, CT; IARC, France;
150
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
University of California - Riverside, CA; Integrated Laboratory Systems, Inc., NC;
California Environmental Protection Agency, CA.)
159. Hamilton-Koch W, Snyder RD, Lavelle JM. 1986. Metal-induced DNA damage and repair
in human diploid fibroblasts and Chinese hamster ovary cells. Chem Biol Interact 59(1):
17-28. (as cited in IARC 2006)
160. Hammond EM, Giaccia AJ. 2005. The role of p53 in hypoxia-induced apoptosis.
Biochemical and Biophysical Research Communications 331(3): 718-725. (Supported by
NIH. Authors affiliated with Stanford University, CA.)
161. Hanna PM, Kadiiska MB, Mason RP. 1992. Oxygen-derived free radical and active oxygen
complex formation from cobalt(II) chelates in vitro. Chem Res Toxicol 5(1): 109-115.
(Support not reported. Authors affiliated with NIEHS, NC; Bulgarian Academy of
Sciences, Bulgaria.)
162. Hansen T, Clermont G, Alves A, Eloy R, Brochhausen C, Boutrand JP, Gatti AM,
Kirkpatrick CJ. 2006. Biological tolerance of different materials in bulk and
nanoparticulate form in a rat model: sarcoma development by nanoparticles. J R Soc
Interface 3(11): 767-775. (Supported by the European Commission. Authors affiliated with
Johannes Gutenberg-University of Mainz, Germany; Biomatech S.A., France; Universita
degli studi di Modena e Reggio Emilia, Italy.)
163. Hartung M, Schaller K-H, Brand E. 1982. On the question of the pathogenic importance of
cobalt for hard metal fibrosis of the lung. Int Arch Occup Environ Health 50: 53-57. (as
cited in IARC 1991)
164. Hartwig A, Kasten U, Boakye-Dankwa K, Schlepegrell R, Beyersmann D. 1990. Uptake
and genotoxicity of micromolar concentrations of cobalt chloride in mammalian cells.
Toxicol Environ Chem 28: 205-215. (as cited in IARC 2006)
165. Hartwig A, Snyder RD, Schlepegrell R, Beyersmann D. 1991. Modulation by Co(II) of
UV-induced DNA repair, mutagenesis and sister-chromatid exchanges in mammalian cells.
Mutat Res 248(1): 177-185. (Supported by the Bundesminister für Forschung und
Technologie, Bonn. Authors affiliated with University of Bremen, Germany; Merrell Dow
Research Institute, OH.)
166. Hartwig A. 1998. Carcinogenicity of metal compounds: possible role of DNA repair
inhibition. Toxicol Lett 102-103: 235-239. (Supported by the Deutsche
Forschungsgemeinschaft. Author affiliated with University of Karlsruhe, germany.)
167. Hartwig A, Asmuss M, Ehleben I, Herzer U, Kostelac D, Pelzer A, Schwerdtle T, Burkle
A. 2002. Interference by toxic metal ions with DNA repair processes and cell cycle control:
molecular mechanisms. Environ Health Perspect 110 Suppl 5: 797-799. (Supported by the
Deutsche Forschungsgemeinschaft, the European Union and Programm Lebensgrundlage
Umwelt und ihre Sicherung (BW-PLUS). Authors affiliated with University of Karlsruhe,
Germany; University of Newcastle, UK; Institut für Strahlenhygiene, Germany.)
This draft document should not be construed to represent final NTP determination or policy
151
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
168. Harty LC, Guinee DG, Travis WD, Bennett WP, Jett J, Colby TV, Tazelaar H, Trastek V,
Pairolero P, Liotta LA, Harris CC, Caporaso NE. 1996. p53 mutations and occupational
exposures in a surgical series of lung cancers. Cancer Epidemiology Biomarkers &
Prevention 5(12): 997-1003. (Support not reported. Authors affiliated with NIH, MD;
Armed Forces Institute of Pathology. Washington. D. C.; Mayo Clinic, MN.)
169. He K. 2011. Trace elements in nails as biomarkers in clinical research. Eur J Clin Invest
41(1): 98-102. (Supported by NIH. Author affiliated with University of North Carolina,
NC.)
170. Heath JC. 1956. The production of malignant tumours by cobalt in the rat. Br J Cancer
10(4): 668-673. (Supported by the British Empire Cancer Campaign. Author affiliated with
Strangeways Research Laboratory, UK.)
171. Heath JC, Daniel MR. 1962. The production of malignant tumours by cobalt in the rat:
intrathoracic tumours. Br J Cancer 16(3): 473-478. (Supported by the British Empire
Cancer Campaign. Author affiliated with Strangeways Research Laboratory, UK.)
172. Heath JC, Webb M. 1967. Content and intracellular distribution of the inducing metal in
the primary rhabdomyosarcomata induced in the rat by cobalt, nickel and cadmium. Br J
Cancer 21(4): 768-779. (Supported by the British Empire Cancer Campaign and the
Medical Research Council. Authors affiliated with Strangeways Research Laboratory, UK.)
173. Heath JC, Webb M, Caffrey M. 1969. The interaction of carcinogenic metals with tissues
and body fluids. Cobalt and horse serum. Br J Cancer 23(1): 153-166. (Supported by the
British Empire Cancer Campaign and the Medical Research Council. Authors affiliated
with Strangeways Research Laboratory, UK.)
174. Henderson RG, Verougstraete V, Anderson K, Arbildua JJ, Brock TO, Brouwers T,
Cappellini D, Delbeke K, Herting G, Hixon G, Odnevall Wallinder I, Rodriguez PH, Van
Assche F, Wilrich P, Oller AR. 2014. Inter-laboratory validation of bioaccessibility testing
for metals. Regul Toxicol Pharmacol 70(1): 170-181. (Supported by the Nickel Producers
Environmental Research Association, Inc., the Cobalt Development Institute, the
International Zinc Association, the European Copper Institute, Eurometaux and the Food
Safety and Environmental Stewardship Program at OSU. Authors affiliated with
ToxStrategies, Inc., NC; Eurometaux, Belgium; Oregon State University, OR; Adolfo
Ibañez University, Chile; Duke University, NC; ECTX bvba, Belgium; Kirby Memorial
Health Center, PA; European Copper Institute, Belgium; KTH Royal Institute of
Technology, Sweden; International Zinc Association, Belgium; Freie Universität Berlin,
Germany; Nickel Producers Environmental Research Association, Inc., NC.)
175. Hengstler JG, Bolm-Audorff U, Faldum A, Janssen K, Reifenrath M, Gotte W, Jung D,
Mayer-Popken O, Fuchs J, Gebhard S, Bienfait HG, Schlink K, Dietrich C, Faust D, Epe B,
Oesch F. 2003. Occupational exposure to heavy metals: DNA damage induction and DNA
repair inhibition prove co-exposures to cadmium, cobalt and lead as more dangerous than
hitherto expected. Carcinogenesis 24(1): 63-73. (Supported by the BMBF and the DFG.
Authors affiliated with University of Mainz, Germany; Ministry of Labour Inspection,
Germany.)
152
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
176. Hennig FF, Raithel HJ, Schaller KH, Dohler JR. 1992. Nickel-, chrom- and cobaltconcentrations in human tissue and body fluids of hip prosthesis patients. J Trace Elem
Electrolytes Health Dis 6(4): 239-243. (as cited in Schaffer et al 1999)
177. Hill AB. 1965. The environment and disease: Association or causation? Proc R Soc Med
58: 295-300. (Support not reported. Author affiliated with University of London, UK.)
178. Hillwalker WE, Anderson KA. 2014. Bioaccessibility of metals in alloys: evaluation of
three surrogate biofluids. Environ Pollut 185: 52-58. (Supported by the OSU Food Safety
and Environmental Stewardship Program. Authors affiliated with Oregon State University,
OR.)
179. Hollins JG, McCullough RS. 1971. Radiation dosimetry of internal contamination by
inorganic compounds of cobalt: an analysis of cobalt metabolism in rats. Health Phys
21(2): 233-246. (Support not reported. Authors affiliated with National Research Council
of Canada, Canada.)
180. Holstein H, Ranebo Y, Raaf CL. 2015. Human metabolism of orally administered
radioactive cobalt chloride. J Environ Radioact 143: 152-158. (Supported by the Swedish
Radiation Safety Authority. Authors affiliated with Skåne University Hospital, Sweden;
Barsebäck Kraft AB, Sweden; Lund University, Sweden.)
181. Hong H-H, Hoenerhoff MJ, Ton T-V, Herbert RA, Kissling GE, Hooth MJ, Behl M, Witt
KL, Smith-Roe SL, Sills RC, Pandiri AR. 2015. Kras, egfr, and tp53 mutations in
B6C3F1/N mouse and F344/NTAC rat alveolar/bronchiolar carcinomas resulting from
chronic inhalation exposure to cobalt metal. Toxicol Pathol(In press). (Supported by the
Divisions of the National Toxicology Program and the Intramural Research Program at the
NIEHS, NIH. Authors affiliated with NIEHS, NC; Experimental Pathology Laboratories,
Inc., NC.)
182. Hopps HC, Stanley AJ, Shideler AM. 1954. Polycythemia induced by cobalt. III.
Histologic studies with evaluation of toxicity of cobaltous chloride. Am J Clin Pathol
24(12): 1374-1380. (Support not reported. Authors affiliated with University of Oklahoma
School of Medicine, OK.)
183. Horev-Azaria L, Kirkpatrick CJ, Korenstein R, Marche PN, Maimon O, Ponti J, Romano
R, Rossi F, Golla-Schindler U, Sommer D, Uboldi C, Unger RE, Villiers C. 2011.
Predictive toxicology of cobalt nanoparticles and ions: comparative in vitro study of
different cellular models using methods of knowledge discovery from data. Toxicol Sci
122(2): 489-501. (Supported by the European Commission. Authors affiliated with Tel
Aviv University, Israel; University Medical Center of the Johannes Gutenberg University
Mainz, Germany; University Grenoble, France; INSERM, France; European Commission,
Italy; Universitaet Muenster, Germany.)
184. Horie M, Fujita K, Kato H, Endoh S, Nishio K, Komaba LK, Nakamura A, Miyauchi A,
Kinugasa S, Hagihara Y, Niki E, Yoshida Y, Iwahashi H. 2012. Association of the physical
and chemical properties and the cytotoxicity of metal oxide nanoparticles: metal ion
release, adsorption ability and specific surface area. Metallomics 4(4): 350-360. (Supported
by the New Energy and Industrial Technology Development Organization of Japan.
This draft document should not be construed to represent final NTP determination or policy
153
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Authors affiliated with University of Occupational and Environmental Health, Japan;
National Institute of Advanced Industrial Science and Technology, Japan; Research
Institute of Science for Safety and Sustainability, Japan; National Metrology Institute of
Japan, Japan; Research Institute for Environmental Management Technology, Japan; Gifu
University, Japan.)
185. HPD. 2014. Household Products Database. National Library of Medicine. Updated on
8/14. http://householdproducts.nlm.nih.gov/advancedsearch.htm and select "Ingredient"
and search CAS No. Accessed on 10/10/14.
186. Huk OL, Catelas I, Mwale F, Antoniou J, Zukor DJ, Petit A. 2004. Induction of apoptosis
and necrosis by metal ions in vitro. J Arthroplasty 19(8 Suppl 3): 84-87. (Supoprted by the
Orthopaedic Research Fund of the Sir Mortimer B. Davis—Jewish General Hospital,
Montreal, Quebec Canada. Authors affiliated with McGill University, Canada.)
187. Hunter DJ, Morris JS, Chute CG, Kushner E, Colditz GA, Stampfer MJ, Speizer FE,
Willett WC. 1990. Predictors of selenium concentration in human toenails. Am J Epidemiol
132(1): 114-122. (Supported by NIH. Authors affiliated with Harvard Medical School,
MA; Brigham and Women's Hospital, MA; University of Missouri, MO.)
188. IARC. 1991. Cobalt and cobalt compounds. In Chlorinated Drinking-water; Chlorination
By-products; Some Other Halogenated Compounds; Cobalt and Cobalt Compounds. IARC
Monographs on the Evaluation of Carcinogenic Risks to Humans, vol. 52. Lyon, France:
International Agency for Research on Cancer. pp. 363-434.
189. IARC. 2006. Metallic cobalt particles (with or without tungsten carbide). In Cobalt in Hard
Metals. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, vol. 86.
Lyon, France: International Agency for Research on Cancer. pp. 39-155.
190. Ichikawa Y, Kusaka Y, Goto S. 1985. Biological monitoring of cobalt exposure, based on
cobalt concentrations in blood and urine. Int Arch Occup Environ Health 55(4): 269-276.
(as cited in IARC 2006 and IARC 1991)
191. Inaba J, Suzuki-Yasumoto M. 1979. A kinetic study of radionuclide absorption through
damaged and undamaged skin of the guinea pig. Health Phys 37(4): 592-595. (Support not
reported. Authors affiliated with National Institute of Radiological Sciences, Japan.)
192. Inoue T, Ohta Y, Sadaie Y, Kada T. 1981. Effect of cobaltous chloride on spontaneous
mutation induction in a Bacillus subtilis mutator strain. Mutat Res 91(1): 41-45. (as cited in
IARC 2006)
193. IPCS. 2015. IPCS INCHEM International Chemical Safety Cards. International
Programme on Chemical Safety. http://www.inchem.org/pages/search.html and search
"International Chemical Safety Cards" by CAS number. Accessed on 3/24/15.
194. Itsara LS, Kennedy SR, Fox EJ, Yu S, Hewitt JJ, Sanchez-Contreras M, Cardozo-Pelaez F,
Pallanck LJ. 2014. Oxidative stress is not a major contributor to somatic mitochondrial
DNA mutations. PLoS Genet 10(2): e1003974. (Supported by a National Institute of
Neurological Disorders and Stroke fellowship, the National Institute of General Medicine,
154
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
the Muscular Dystrophy Association, the Nathan Shock Center of Excellence in Basic
Biology of Aging, the National Cancer Institute and the National Institute on Aging.
Authors affiliated with University of Washington, WA; Mayo Clinic, FL; University of
Montana, MT.)
195. Jansen HM, Knollema S, van der Duin LV, Willemsen AT, Wiersma A, Franssen EJ,
Russel FG, Korf J, Paans AM. 1996. Pharmacokinetics and dosimetry of cobalt-55 and
cobalt-57. J Nucl Med 37(12): 2082-2086. (Supported by the Dutch Technology
Foundation. Authors affiliated with University of Groningen, Netherlands; University of
Nijmegen, Netherlands.)
196. Jantzen C, Jorgensen HL, Duus BR, Sporring SL, Lauritzen JB. 2013. Chromium and
cobalt ion concentrations in blood and serum following various types of metal-on-metal hip
arthroplasties: a literature overview. Acta Orthop 84(3): 229-236. (Support not reported.
Authors affiliated with University of Copenhagen, Denmark.)
197. Jasmin G, Riopelle JL. 1976. Renal carcinomas and erythrocytosis in rats following
intrarenal injection of nickel subsulfide. Lab Invest 35(1): 71-78. (Supported by the
Medical Research Council of Canada. Authors affiliated with Universite de Montreal,
Canada.)
198. Jemal A, Simard EP, Xu J, Ma J, Anderson RN. 2013. Selected cancers with increasing
mortality rates by educational attainment in 26 states in the United States, 1993-2007.
Cancer Causes Control 24(3): 559-565. (Supported by the Intramural Research Program of
the American Cancer Society. Authors affiliated with American Cancer Society, GA;
Centers for Disease Control and Prevention, MD.)
199. Jiang H, Liu F, Yang H, Li Y. 2012. Effects of cobalt nanoparticles on human T cells in
vitro. Biol Trace Elem Res 146(1): 23-29. (Supported by the National Natural Science
Foundation of China. Authors affiliated with First Affiliated Hospital of Soochow
University, China; Affiliated Hospital of Nantong University, China; First People’s
Hospital of Taizhou City, China.)
200. Jomova K, Valko M. 2011. Advances in metal-induced oxidative stress and human disease.
Toxicology 283(2-3): 65-87. (Supported by the Scientific Grant Agency of the Slovak
Republic and the Slovak Research and Development Agency of the Slovak Republic.
Authors affiliated with Constantine The Philosopher University, Slovakia; Slovak
Technical University, Slovakia.)
201. Kada T, Kanematsu N. 1978. Reduction of N -methyl-N’-nitrosoguanidine-induced
mutations by cobalt chloride in Escherichia coli. Proc Jpn Acad 54B: 234-237. (as cited in
IARC 2006)
202. Kadiiska MB, Maples KR, Mason RP. 1989. A comparison of cobalt(II) and iron(II)
hydroxyl and superoxide free radical formation. Arch Biochem Biophys 275(1): 98-111.
(Support not reported. Authors affiliated with NIEHS, NC.)
203. Kain J, Karlsson HL, Moller L. 2012. DNA damage induced by micro- and nanoparticles-interaction with FPG influences the detection of DNA oxidation in the comet assay.
This draft document should not be construed to represent final NTP determination or policy
155
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Mutagenesis 27(4): 491-500. (Supported by the Swedish Research Council for
Environment, Agricultural Sciences and Spatial Planning, the Swedish Research Council,
AF’s Foundation for Research and Development. Authors affiliated with Karolinska
Institutet, Sweden.)
204. Kanabrocki EL, Kanabrocki JA, Greco J, Kaplan E, Oester YT, Brar SS, Gustafson PS,
Nelson DM, Moore CE. 1979. Instrumental analysis of trace elements in thumbnails of
human subjects. Sci Total Environ 13(2): 131-140. (Supported by WHO. Authors affiliated
with Veterans Administration Hospital, IL; Argonne National Laboratory, IL; Loyola
University, IL.)
205. Kanaji A, Orhue V, Caicedo MS, Virdi AS, Sumner DR, Hallab NJ, Yoshiaki T, Sena K.
2014. Cytotoxic effects of cobalt and nickel ions on osteocytes in vitro. J Orthop Surg Res
9(1): 91. (Support not reported. Authors affiliated with Rush University Medical Center,
IL; Keio University, Japan; Kagoshima University, Japan.)
206. Kanematsu N, Hara M, Kada T. 1980. Rec assay and mutagenicity studies on metal
compounds. Mutat Res 77(2): 109-116. (as cited in IARC 2006)
207. Kanias GD, Kouri E, Arvaniti H, Karaiosifidi H, Kouneli S. 1994. Trace element content in
breasts with fibrocystic disease. Biol Trace Elem Res 43-45: 363-370. (Support not
reported. Authors affiliated with N.C.S.R. DEMOKRITOS, Greece; "Helena Venizelou,"
Greece. )
208. Kasirsky G, Gautieri RF, Mann DE, Jr. 1965. Effect of cobaltous chloride on the minimal
carcinogenic dose of methylcholanthrene in albino mice. J Pharm Sci 54: 491-493.
(Support and author affiliations not reported.)
209. Kasprzak KS, Zastawny TH, North SL, Riggs CW, Diwan BA, Rice JM, Dizdaroglu M.
1994. Oxidative DNA base damage in renal, hepatic, and pulmonary chromatin of rats after
intraperitoneal injection of cobalt(II) acetate. Chem Res Toxicol 7(3): 329-335. (Support
not reported. Authors affiliated with NCI, MD; Data Management Services, Inc., MD;
NIST, MD; Rydygier Medical School, Poland; Program hources, Inc./DynCorp, MD.)
210. Kasprzak KS. 2002. Oxidative DNA and protein damage in metal-induced toxicity and
carcinogenesis. Free Radical Biology and Medicine 32(10): 958-967. (Support not
reported. Authors affiliated with NCI, MD.)
211. Kasten U, Mullenders LH, Hartwig A. 1997. Cobalt(II) inhibits the incision and the
polymerization step of nucleotide excision repair in human fibroblasts. Mutat Res 383(1):
81-89. (Supported by the Kooperationsbereich Universität/Arbeiterkammer Bremen and by
the Deutsche Forschungsgemeinschaft. Authors affiliated with University of Bremen,
Germany; Leiden University, Netherlands.)
212. Kawanishi S, Inoue S, Yamamoto K. 1989. Hydroxyl radical and singlet oxygen
production and DNA damage induced by carcinogenic metal compounds and hydrogen
peroxide. Biol Trace Elem Res 21: 367-372. (Support not reported. Authors affiliated with
Kyoto University, Japan.)
156
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
213. Kawanishi S, Inoue S, Yamamoto K. 1994. Active oxygen species in DNA damage
induced by carcinogenic metal compounds. Environ Health Perspect 102 Suppl 3: 17-20.
(Support not reported. Authors affiliated with Kyoto University, Japan.)
214. Keegan GM, Learmonth ID, Case CP. 2008. A systematic comparison of the actual,
potential, and theoretical health effects of cobalt and chromium exposures from industry
and surgical implants. Crit Rev Toxicol 38(8): 645-674. (Support not reported. Authors
affiliated with University of Bristol, UK; Southmead Hospital, UK.)
215. Kennedy SM, Chan-Yeung M, Marion S, Lea J, Teschke K. 1995. Maintenance of stellite
and tungsten carbide saw tips: respiratory health and exposure-response evaluations. Occup
Environ Med 52(3): 185-191. (Support not reported. Authors affiliated with University of
British Columbia, Canada.)
216. Kerckaert GA, LeBoeuf RA, Isfort RJ. 1996. Use of the Syrian hamster embryo cell
transformation assay for determining the carcinogenic potential of heavy metal compounds.
Fundam Appl Toxicol 34(1): 67-72. (Support not reported. Authors affiliated with Procter
& Gamble Company, OH.)
217. Kerfoot EJ, Fredrick WG, Domeier E. 1975. Cobalt metal inhalation studies on miniature
swine. Am Ind Hyg Assoc J 36(1): 17-25. (Supported by the Health Services and Mental
Health Administration, Department of Health, Education and Welfare. Authors affiliated
with Wayne State University, MI.)
218. Kharab P, Singh I. 1985. Genotoxic effects of potassium dichromate, sodium arsenite,
cobalt chloride and lead nitrate in diploid yeast. Mutat Res 155(3): 117-120. (as cited in
IARC 2006)
219. Kharab P, Singh I. 1987. Induction of respiratory deficiency in yeast by salts of chromium,
arsenic, cobalt and lead. Indian J Exp Biol 25(2): 141-142. (as cited in IARC 2006)
220. Kibblewhite MG, Van Rensburg SJ, Laker MC, Rose EF. 1984. Evidence for an intimate
geochemical factor in the etiology of esophageal cancer. Environ Res 33(2): 370-378.
(Supported by the South African National Cancer Association. Authors affiliated with
University of Fort Hare, South Africa; National Research Institute for Nutritional Diseases,
South Africa.)
221. Kitahara J, Yamanaka K, Kato K, Lee YW, Klein CB, Costa M. 1996. Mutagenicity of
cobalt and reactive oxygen producers. Mutat Res 370(3-4): 133-140. (Supported by NIH.
Authors affiliated with New York University Medical Center, NY; Nihon University
College of Pharmacy, Japan.)
222. Klatka J, Remer M, Dobrowolski R, Pietruszewska W, Trojanowska A, Siwiec H,
Charytanowicz M. 2011. The content of cadmium, cobalt and nickel in laryngeal
carcinoma. Arch Med Sci 7(3): 517-522. (Support not reported. Authors affiliated with
Medical University, Poland; District Hospital, Poland; Maria Curie Sklodowska University,
Poland; The John Paul II Catholic University, Poland.)
This draft document should not be construed to represent final NTP determination or policy
157
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
223. Klaunig JE, Kamendulis LM, Hocevar BA. 2010. Oxidative stress and oxidative damage in
carcinogenesis. Toxicol Pathol 38(1): 96-109. (Support not reported. Authors affiliated
with Indiana University School of Medicine, IN.)
224. Klein CB, Su L, Rossman TG, Snow ET. 1994. Transgenic gpt+ V79 cell lines differ in
their mutagenic response to clastogens. Mutat Res 304(2): 217-228. (Supported by NIH and
NIEHS. Authors affiliated with New York University Medical Center, NY.)
225. Koedrith P, Seo YR. 2011. Advances in carcinogenic metal toxicity and potential
molecular markers. Int J Mol Sci 12(12): 9576-9595. (Supported by the Ministry of
Environment, Korea, the Korea Research Foundation, the National Research Foundation of
Korea and the Ministry of Education, Science and Technology, Korea. Authors affiliated
with Dongguk University, Korea.)
226. Kopera E, Schwerdtle T, Hartwig A, Bal W. 2004. Co(II) and Cd(II) substitute for Zn(II) in
the zinc finger derived from the DNA repair protein XPA, demonstrating a variety of
potential mechanisms of toxicity. Chem Res Toxicol 17(11): 1452-1458. (Supported by the
Polish State Committee for Scientific Research and the Deutsche Forschungsgemeinschaft.
Authors affiliated with Polish Academy of Sciences, Poland; Technical University of
Berlin, Germany.)
227. Kotake-Nara E, Saida K. 2007. Characterization of CoCl2-induced reactive oxygen species
(ROS): Inductions of neurite outgrowth and endothelin-2/vasoactive intestinal contractor in
PC12 cells by CoCl2 are ROS dependent, but those by MnCl2 are not. Neurosci Lett
422(3): 223-227. (Supported by a New Energy and Industrial Technology Development
Organization (NEDO) fellowship. Authors affiliated with National Institute of Advanced
Industrial Science and Technology, Japan; New Energy and Industrial Technology
Development Organization (NEDO), Japan.)
228. Kraus T, Schramel P, Schaller KH, Zöbelein P, Weber A, Angerer J. 2001. Exposure
assessment in the hard metal manufacturing industry with special regard to tungsten and its
compounds. Occup Environ Med 58(10): 631-634. (Support not reported. Authors affiliated
with University of Erlangen-Nürnberg, Germany; GSF Neuherberg, Germany.)
229. Kreyling WG, Ferron GA, Haider B. 1986. Metabolic fate of inhaled Co aerosols in beagle
dogs. Health Phys 51(6): 773-795. (Support not reported. Authors affiliated with Institut
fuer Stahlenschutz, Germany.)
230. Kreyling WG, Andre S, Collier CG, Ferron GA, Metivier H, Schumann G. 1991a.
Interspecies comparison of lung clearance after inhalation of monodisperse, solid cobalt
oxide aerosol particles. J Aerosol Sci 22(4): 509-535. (Supported by CEC. Authors
affiliated with Gesellschaft fur Strahlen- und Umweltforschung mbH Munchen (GSF),
Germany; Commissariat à l'Energie Atomique (CEA), France; National Radiological
Protection Board (NRPB), UK.)
231. Kreyling WG, Nyberg K, Nolibe D, Collier CG, Camner P, Heilman P, Lirsac PN,
Lundborg M, Matejkova E. 1991b. Interspecies comparison of phagolysosomal pH in
alveolar macrophages. Inhal Toxicol 3: 91-100. (Supported by the European Late Effects
Project Group, the CEC, and the Karolinska Intitute. Authors affiliated with Gesellschaft
158
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
für Strahlen- und Umweltforschung mbH München, Germany; Karolinska Institutet,
Sweden; Commissariat a I'Energie Atomique, France; National Radiological Protection
Board, UK.)
232. Kubo A, Levin TR, Block G, Rumore GJ, Quesenberry CP, Jr., Buffler P, Corley DA.
2009. Alcohol types and sociodemographic characteristics as risk factors for Barrett's
esophagus. Gastroenterology 136(3): 806-815. (Supported by NIH and Kaiser
Permanente/Permanente Medical Group Community Benefits. Authors affiliated with
Kaiser Permanente, CA; University of California - Berkeley and San Francisco, CA.)
233. Kuo CY, Wong RH, Lin JY, Lai JC, Lee H. 2006. Accumulation of chromium and nickel
metals in lung tumors from lung cancer patients in Taiwan. J Toxicol Environ Health A
69(14): 1337-1344. (Support not reported. Authors affiliated with Chung Shan Medical
University, China.)
234. Kusaka Y. 1996. Cobalt and nickel induced hard metal asthma. Toxicol Metals: 461-468.
(Support and author affiliations not reported.)
235. Kusama T, Itoh S, Yoshizawa Y. 1986. Absorption of radionuclides through wounded skin.
Health Phys 51(1): 138-141. (Support not reported. Authors affiliated with University of
Tokyo, Japan.)
236. Kwon YM, Xia Z, Glyn-Jones S, Beard D, Gill HS, Murray DW. 2009. Dose-dependent
cytotoxicity of clinically relevant cobalt nanoparticles and ions on macrophages in vitro.
Biomed Mater 4(2): 025018. (Support not reported. Authors affiliated with University of
Oxford, UK.)
237. Kyono H, Kusaka Y, Homma K, Kubota H, Endo-Ichikawa Y. 1992. Reversible lung
lesions in rats due to short-term exposure to ultrafine cobalt particles. Ind Health 30(3-4):
103-118. (Support not reported. Authors affiliated with National Institute of Industrial
Health, Japan; Fukui Medical School, Japan; Kansai Medical University, Japan.)
238. Lacy SA, Merritt K, Brown SA, Puryear A. 1996. Distribution of nickel and cobalt
following dermal and systemic administration with in vitro and in vivo studies. J Biomed
Mater Res 32(2): 279-283. (Support not reported. Authors affiliated with Case Western
Reserve University, OH.)
239. Laker M. 1982. On determining trace element levels in man: the uses of blood and hair.
Lancet 2(8292): 260-262. (Support not reported. Authors affiliated with Royal Hospital for
Sick Children, UK.)
240. Larese Filon F, Maina G, Adami G, Venier M, Coceani N, Bussani R, Massiccio M,
Barbieri P, Spinelli P. 2004. In vitro percutaneous absorption of cobalt. Int Arch Occup
Environ Health 77(2): 85-89. (Supported by the European Community. Authors affiliated
with Università di Trieste, Italy; Università degli Studi di Torino, Italy; Eurand, Italy.)
241. Larese Filon F, Gianpietro A, Venier M, Maina G, Renzi N. 2007. In vitro percutaneous
absorption of metal compounds. Toxicol Lett 170(1): 49-56. (Supported by the European
This draft document should not be construed to represent final NTP determination or policy
159
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Community. Authors affiliated with Scienze di Medicina Pubblica, Italy; University of
Trieste, Italy.)
242. Larese Filon F, D'Agostin F, Crosera M, Adami G, Bovenzi M, Maina G. 2009. In vitro
absorption of metal powders through intact and damaged human skin. Toxicol In Vitro
23(4): 574-579. (Supported by the Italian Minister of University and Research. Authors
affiliated with Università di Trieste, Italy.)
243. Lark RM, Ander EL, Cave MR, Knights KV, Glennon MM, Scanlon RP. 2013. Mapping
trace element deficiency by cokriging from regional geochemical soil data: A case study on
cobalt for grazing sheep in Ireland. Geoderma 226-227: 64-78. (Supported by the European
Union's INTERREG IVA cross-border Programme managed by the Special EU
Programmes Body, the Tellus Border Project TB: ‘Application of the Tellus Border soil
chemistry data to the agricultural sector in Ireland,’ and NERC. Authors affiliated with
British Geological Survey, UK; Geological Survey of Ireland, Ireland.)
244. Lee JC, Son YO, Pratheeshkumar P, Shi X. 2012. Oxidative stress and metal
carcinogenesis. Free Radical Biology and Medicine 53(4): 742-757. (Supported by NIH.
Authors affiliated with University of Kentucky, KY; Chonbuk National University, South
Korea.)
245. Lee SG, Lee H, Rho HM. 2001. Transcriptional repression of the human p53 gene by
cobalt chloride mimicking hypoxia. FEBS Lett 507(3): 259-263. (Supported by Research
Grants from Molecular Medicine Research Program of the Ministry of Science and
Technology (MOST) and Research Fellowships from the Ministry of Education and
Human Resources Development. Authors affiliated with Seoul National University, South
Korea.)
246. Leggett RW. 2008. The biokinetics of inorganic cobalt in the human body. Sci Total
Environ 389(2-3): 259-269. (Supported by the Office of Radiation and Indoor Air, U. S.
Environmental Protection Agency. Author affiliated with Oak Ridge National Laboratory,
TN.)
247. Leitao AC, Soares RA, Cardoso JS, Guillobel HC, Caldas LR. 1993. Inhibition and
induction of SOS responses in Escherichia coli by cobaltous chloride. Mutat Res 286(2):
173-180. (as cited in IARC 2006)
248. Leonard S, Gannett PM, Rojanasakul Y, Schwegler-Berry D, Castranova V, Vallyathan V,
Shi X. 1998. Cobalt-mediated generation of reactive oxygen species and its possible
mechanism. J Inorg Biochem 70(3-4): 239-244. (Support not reported. Authors affiliated
with NIOSH; West Virginia University, WV.)
249. Leonard SS, Harris GK, Shi X. 2004. Metal-induced oxidative stress and signal
transduction. Free Radic Biol Med 37(12): 1921-1942. (Support not reported. Authors
affiliated with NIOSH, WV.)
250. Letourneau EG, Jack GC, McCullough RS, Hollins JG. 1972. The metabolism of cobalt by
the normal human male: whole body retention and radiation dosimetry. Health Phys 22(5):
160
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
451-459. (Support not reported. Authors affiliated with Department of National Health and
Welfare, Canada; National Research Council of Canada, Canada.)
251. Lewis CP, Demedts M, Nemery B. 1991. Indices of oxidative stress in hamster lung
following exposure to cobalt(II) ions: in vivo and in vitro studies. Am J Respir Cell Mol
Biol 5(2): 163-169. (Supported by the Katholieke Universiteit Leuven and the Fonds voor
Geneeskundig Wetenschappelijk Onderzoek (FGWO), Brussels. Authors affiliated with
Katholieke Universiteit Leuven, Belgium.)
252. Lewis CP, Demedts M, Nemery B. 1992. The role of thiol oxidation in Cobalt(II)-induced
toxicity in hamster lung. Biochem Pharmacol 43(3): 519-525. (Supported by the European
Science Foundation, the Belgian Fonds voor Geneeskundig Wetenschappelijk Onderzoek
and by the Onderzoeksfonds of the Katholieke Universiteit Leuven. Authors affiliated with
Katholieke Universiteit Leuven, Belgium.)
253. Limbach LK, Wick P, Manser P, Grass RN, Bruinink A, Stark WJ. 2007. Exposure of
engineered nanoparticles to human lung epithelial cells: influence of chemical composition
and catalytic activity on oxidative stress. Environ Sci Technol 41(11): 4158-4163.
(Supported by the Swiss Federal Office of Public Health. Authors affiliated with ETH
Zurich, Switzerland; Swiss Federal Laboratories for Materials Testing and Research,
Switzerland.)
254. Linna A, Oksa P, Palmroos P, Roto P, Laippala P, Uitti J. 2003. Respiratory health of
cobalt production workers. Am J Ind Med 44(2): 124-132. (Supported by the Finnish Work
Environment Fund. Authors affiliated with Outokumpu Zinc Oy, Finland; Tampere
University Hospital, Finland.)
255. Linnainmaa M, Kiilunen M. 1997. Urinary cobalt as a measure of exposure in the wet
sharpening of hard metal and stellite blades. Int Arch Occup Environ Health 69(3): 193200. (Supported by the Finnish Work Environment Fund. Authors affiliated with Finnish
Institute of Occupational Health, Finland.)
256. Lippi G, Franchini M, Guidi GC. 2005. Cobalt chloride administration in athletes: a new
perspective in blood doping? Br J Sports Med 39(11): 872-873. (Support not reported.
Authors affiliated with Università di Verona, Italy; Azienda Ospedaliera di Verona, Italy.)
257. Lison D, De Boeck M, Verougstraete V, Kirsch-Volders M. 2001. Update on the
genotoxicity and carcinogenicity of cobalt compounds. Occup Environ Med 58(10): 619625. (Support not reported. Authors affiliated with Universite Catholique de Louvain,
Belgium; Vrije Universiteit Brussel, Belgium.)
258. Lison D. 2015. Cobalt. In Handbook on the Toxicology of Metals. 4th ed., Vol. II: Specific
Metals. Nordberg GF, Fowler BA, Nordberg M, eds. Waltham, MA: Elsevier. pp. 743-763.
259. Liu FC, Qin J, Wu HS, Wu YL, Zhu YL. 2011. Co and Cr accumulation in hair after metalon-metal hip resurfacing arthroplasty. ANZ J Surg 81(6): 436-439. (No funding received.
Authors affiliated with Second Military Medical University, China.)
This draft document should not be construed to represent final NTP determination or policy
161
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
260. Lloyd DR, Carmichael PL, Phillips DH. 1998. Comparison of the formation of 8-hydroxy2'-deoxyguanosine and single- and double-strand breaks in DNA mediated by fenton
reactions. Chem Res Toxicol 11(5): 420-427. (Support not reported. Authors affiliated with
Haddow Laboratories, UK; Imperial College School of Medicine at St. Mary’s, Norfolk
Place, UK.)
261. Luyts K, Napierska D, Nemery B, Hoet PH. 2013. How physico-chemical characteristics of
nanoparticles cause their toxicity: complex and unresolved interrelations. Environ Sci
Process Impacts 15(1): 23-38. (Supported by the ENPRA (Risk assessment of engineered
nano particles), Fonds voor Wetenschappelijk Onderzoek Vlaanderen (FWO), KU Leuven
and an agentschap voor Innovatie door Wetenschap en Technologie. Authors affiliated with
K.U. Leuven, Belgium.)
262. Magaye R, Zhao J, Bowman L, Ding M. 2012. Genotoxicity and carcinogenicity of cobalt-,
nickel- and copper-based nanoparticles. Exp Ther Med 4(4): 551-561. (Supported by the
Ningbo Scientific Project, the Scientific Innovation Team Project of Ningbo, the
Foundations of Innovative Research Team of Educational Commission of Zhejiang
Province, Innovative Research Team of Ningbo and K.C. Wong Magna Fund in Ningbo
University. Authors affiliated with Ningbo University, China; NIOSH, WV.)
263. Malard V, Berenguer F, Prat O, Ruat S, Steinmetz G, Quemeneur E. 2007. Global gene
expression profiling in human lung cells exposed to cobalt. BMC Genomics 8. (Supported
by "Electricité De France". Authors affiliated with DSV/iBEB, CEA VALRHO, France.)
264. Mao Y, Liu KJ, Jiang JJ, Shi X. 1996. Generation of reactive oxygen species by Co(II)
from H2O2 in the presence of chelators in relation to DNA damage and 2'-deoxyguanosine
hydroxylation. J Toxicol Environ Health 47(1): 61-75. (Support not reported. Authors
affiliated with NCI, MD; Dartmouth Medical School, NH.)
265. Mates JM, Segura JA, Alonso FJ, Marquez J. 2010. Roles of dioxins and heavy metals in
cancer and neurological diseases using ROS-mediated mechanisms. Free Radic Biol Med
49(9): 1328-1341. (Supported by the Ministerio de Ciencia y Tecnología of Spain and
Junta de Andalucía, Proyectos de Investigación de Excelencia, Convocatoria 2006, Spain.
Authors affiliated with Universidad de Málaga, Spain.)
266. Maxwell P, Salnikow K. 2004. HIF-1: an oxygen and metal responsive transcription factor.
Cancer Biol Ther 3(1): 29-35. (Support not reported. Authors affiliated with Imperial
College, UK; NCI, MD.)
267. Mayo Clinic. 2015. Test ID: COS. Cobalt, serum. Mayo Medical Laboratories.
http://www.mayomedicallaboratories.com/test-catalog/Clinical+and+Interpretive/80084.
Accessed on 4/10/15.
268. McInnes EF, Ernst H, Germann PG. 2013. Spontaneous neoplastic lesions in control Syrian
hamsters in 6-, 12-, and 24-month short-term and carcinogenicity studies. Toxicol Pathol
41(1): 86-97. (No funding received. Authors affiliated with Healthscope, Australia;
Fraunhofer Institute for Toxicology and Experimental Medicine, Germany; Nycomed
GmbH, Germany.)
162
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
269. McKinley JM, Ofterdinger U, Young M, Barsby A, Gavin A. 2013. Investigating local
relationships between trace elements in soils and cancer data. Spatial Stat 5: 25-41.
(Supported by the British Geological Survey (BGS) University Funding Initiative, the
Public Health Agency for N. Ireland, the Department for Enterprise, Trade and Investment
(DETINI) and The Rural Development Programme through the Northern Ireland
Programme for Building Sustainable Prosperity. Authors affiliated with Queen’s University
Belfast, UK; Geological Survey Northern Ireland, UK; N. Ireland Cancer Registry, UK.)
270. McLean JR, McWilliams RS, Kaplan JG, Birnboim HC. 1982. Rapid detection of DNA
strand breaks in human peripheral blood cells and animal organs following treatment with
physical and chemical agents. In Progress in Mutation Research. vol. 3. Bora KC, Douglas
GR, Nestmann ER, eds. Amsterdam: Elsevier Biomedical Press. pp. 137-141. (as cited in
IARC 2006)
271. Meltzer HM, Brantsaeter AL, Borch-Iohnsen B, Ellingsen DG, Alexander J, Thomassen Y,
Stigum H, Ydersbond TA. 2010. Low iron stores are related to higher blood concentrations
of manganese, cobalt and cadmium in non-smoking, Norwegian women in the HUNT 2
study. Environ Res 110(5): 497-504. (Supported by the Norwegian Research Council.
TheNord-TrøndelagHealth Study (The HUNT Study)is a collaboration between HUNT
Research Centre, Faculty of Medicine, Norwegian University of Science and
Technology(NTNU), Verdal, Norwegian Institute of Public Health, and Nord-Trøndelag
County Council. Authors affiliated with Norwegian Institute of Public Health, Norway;
University of Oslo, Norway; National Institute of Occupational Health, Norway;
Norwegian Institute of Public Health, Norway; Statistics Norway, Norway.)
272. Menzel DB, Wolpert RL, Francovitch RJ, Shoaf CR, Boger III JR, Tayyeb MI. 1989.
Respiratory tract burdens of cobalt from inhalation of soluble aerosols: simulation by a
two-compartment model. Inhal Toxicol 1: 49-69. (Supported by NIH and the International
Life Sciences Risk Research Institute. Authors affiliated with Duke University Medical
Center, NC.)
273. Méplan C, Richard MJ, Hainaut P. 2000. Metalloregulation of the tumor suppressor protein
p53: zinc mediates the renaturation of p53 after exposure to metal chelators in vitro and in
intact cells. Oncogene 19(46): 5227-5236. (Supported by the Fondation Volvic pour le
recherche sur les oligoéléments (1998) and by an IARC Special Training Award. Authors
affiliated with IARC, France; Centre Hospitalier et Universitaire Albert Michalon, France.)
274. MHRA. 2012. Medical Device Alert: All Metal-on-Metal (MoM) Hip Replacements. Ref:
MDA/2012/036. Medicines and Healthcare Products Regulatory Agency. 7 pp.
https://assets.digital.cabinetoffice.gov.uk/media/5485abf6ed915d4c10000273/con155767.pdf
275. Mikkelsen S, Raffn E, Altman D, Groth S, Christensen JM. 1984. [Helbred og Kobolt]. In
Tvaersnitsundersøgelse af Plattemalere [Health and Cobalt. A Cross-sectional Study of
Plate Painters]. Copenhagen: Arbejdsmitjøfondet. (as cited in IARC 1991)
276. Miller AC, Mog S, McKinney L, Luo L, Allen J, Xu J, Page N. 2001. Neoplastic
transformation of human osteoblast cells to the tumorigenic phenotype by heavy metal-
This draft document should not be construed to represent final NTP determination or policy
163
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
tungsten alloy particles: induction of genotoxic effects. Carcinogenesis 22(1): 115-125.
(Supported by the Armed Forces Radiobiology Research Institute. Authors affiliated with
Armed Forces Radiobiology Research Institute, MD; NIH, MD.)
277. Mining Technology Market and Customer Insight. 2015. Idaho Cobalt Mine, United States
of America. Mining Technology Market & Customer Insight. http://www.miningtechnology.com/projects/idaho-cobalt-mine/. Accessed on 3/6/15.
278. Miyaki M, Akamatsu N, Ono T, Koyama H. 1979. Mutagenicity of metal cations in
cultured cells from Chinese hamster. Mutat Res 68(3): 259-263. (as cited in IARC 2006)
279. Mo Y, Zhu X, Hu X, Tollerud DJ, Zhang Q. 2008. Cytokine and NO release from
peripheral blood neutrophils after exposure to metal nanoparticles: In vitro and ex vivo
studies. Nanotoxicology 2(2): 79-87. (Supported by the American Lung Association and the
University of Louisville. Authors affiliated with University of Louisville, KY; Zhejiang
University, China; Albert Einstein College of Medicine, NY.)
280. Mochizuki H, Kada T. 1982. Antimutagenic action of cobaltous chloride on Trp-P-1induced mutations in Salmonella typhimurium TA98 and TA1538. Mutat Res 95(2-3): 145157. (as cited in IARC 2006)
281. Mohmand J, Eqani SA, Fasola M, Alamdar A, Mustafa I, Ali N, Liu L, Peng S, Shen H.
2015. Human exposure to toxic metals via contaminated dust: Bio-accumulation trends and
their potential risk estimation. Chemosphere 132: 142-151. (Supported by the Higher
Education Commission and the the IUE, CAS, China. Authors affiliated with COMSATS
Institute of Information & Technology, Pakistan; Università, Italy; Chinese Academy of
Sciences, China; University of Sargodha, Pakistan; King Abdulaziz University, Saudi
Arabia.)
282. Moorhouse CP, Halliwell B, Grootveld M, Gutteridge JM. 1985. Cobalt(II) ion as a
promoter of hydroxyl radical and possible 'crypto-hydroxyl' radical formation under
physiological conditions. Differential effects of hydroxyl radical scavengers. Biochim
Biophys Acta 843(3): 261-268. (Supported by the Wellcome Trust and the Arthritis and
Rheumatism Council. Authors affiliated with University of London King's College, UK;
National Institute for Biological Standards and Control, UK.)
283. Moreno ME, Acosta-Saavedra LC, Meza-Figueroa D, Vera E, Cebrian ME, OstroskyWegman P, Calderon-Aranda ES. 2010. Biomonitoring of metal in children living in a
mine tailings zone in Southern Mexico: A pilot study. Int J Hyg Environ Health 213(4):
252-258. (Supported by the Mexican Council for Science and Technology. Authors
affiliated with Cinvestav, Mexico; Universidad Autonoma de Guerrero, Mexico;
Universidad de Sonora, Mexico; Universidad Nacional Autonoma de Mexico, Mexico.)
284. Mosconi G, Bacis M, Vitali MT, Leghissa P, Sabbioni E. 1994a. Cobalt excretion in urine:
results of a study on workers producing diamond grinding tools and on a control group. Sci
Total Environ 150(1-3): 133-139. (Support not reported. Authors affiliated with Ospedali
Riuniti of Bergamo, Italy; Commission of the European Communities, Italy.)
164
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
285. Mosconi G, Bacis M, Leghissa P, Maccarana G, Arsuffi E, Imbrogno P, Airoldi L, Caironi
M, Ravasio G, Parigi PC, Polini S, Luzzana G. 1994b. Occupational exposure to metallic
cobalt in the Province of Bergamo. Results of a 1991 survey. Sci Total Environ 150(1-3):
121-128. (Support not reported. Authors affiliated with Ospedali Riuniti of Bergamo, Italy;
Local Sanitary Units 26 - 32, Italy.)
286. Moulin JJ, Wild P, Mur JM, Fournier-Betz M, Mercier-Gallay M. 1993. A mortality study
of cobalt production workers: an extension of the follow-up. Am J Ind Med 23(2): 281-288.
(Support not reported. Authors affiliated with INRS, France.)
287. Moulin JJ, Wild P, Romazini S, Lasfargues G, Peltier A, Bozec C, Deguerry P, Pellet F,
Perdrix A. 1998. Lung cancer risk in hard-metal workers. Am J Epidemiol 148(3): 241-248.
(Support not reported. Authors affiliated with INRS, France; Institut Universtiaire de
Medecine du Travail et d'Ergonomie, France; Institut Universtiaire de Medecine du Travail
du Val de Loire, France; Eramet, France; Sandvik Hard Materials, France.)
288. Moulin JJ, Clavel T, Roy D, Dananché B, Marquis N, Févotte J, Fontana JM. 2000a. Risk
of lung cancer in workers producing stainless steel and metallic alloys. Int Arch Occup
Environ Health 73(3): 171-180. (Support not reported. Authors affiliated with INRS,
France; Paris La Défense, France; Institut Universitaire de Médecine du Travail de Lyon,
France; LECES, France.)
289. Moulin JJ, Clavel T, Buclez B, Laffitte-Rigaud G. 2000b. A mortality study among
workers in a French aluminium reduction plant. Int Arch Occup Environ Health 73(5): 323330.
290. Munakata M, Onuma A, Kobayashi Y, Haginoya K, Yokoyama H, Fujiwara I, Yasuda H,
Tsutsui T, Iinuma K. 2006. A preliminary analysis of trace elements in the scalp hair of
patients with severe motor disabilities receiving enteral nutrition. Brain Dev 28(8): 521525.
291. Mur JM, Moulin JJ, Charruyer-Seinerra MP, Lafitte J. 1987. A cohort mortality study
among cobalt and sodium workers in an electrochemical plant. Am J Ind Med 11(1): 75-81.
(Support not reported. Authors affiliated with INRS, France.)
292. Naylor GP, Harrison JD. 1995. Gastrointestinal iron and cobalt absorption and iron status
in young rats and guinea pigs. Hum Exp Toxicol 14(12): 949-954. (Support not reported.
Authors affiliated with National Radiological Protection Board, UK.)
293. Neale G. 1990. B12 binding proteins. Gut 31(1): 59-63. (Support not reported. Author
affiliated with Addenbrookes Hospital, UK.)
294. Newton D, Rundo J. 1971. The long-term retention of inhaled cobalt-60. Health Phys
21(3): 377-384. (Support not reported. Authors affiliated with Atomic Energy Research
Establishment, UK; Argonne National Laboratory, IL.)
295. Nielsen NH, Kristiansen J, Borg L, Christensen JM, Poulsen LK, Menne T. 2000. Repeated
exposures to cobalt or chromate on the hands of patients with hand eczema and contact
allergy to that metal. Contact Dermatitis 43(4): 212-215. (Supported by the European
This draft document should not be construed to represent final NTP determination or policy
165
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Commission. Authors affiliated with University of Copenhagen, Denmark; National
Institute of Occupational Health, Denmark; National University Hospital, Denmark.)
296. Nilsson K, Jensen BS, Carlsen L. 1985. The migration chemistry of cobalt. Eur Appl Res
Rept - Nucl Sci Technol 7(1): 23-86. (Performed in the frame of the R&D Programme
"Management and disposal of radioactive waste." Authors affiliated with Risø National
Laboratory, Denmark.)
297. NIOSH. 1987b. Health Hazard Evaluation Report: GTE/Valenite Corporation,
Westminster, South Carolina. HETA 85-064-1844. U.S. Department of Health and Human
Services, National Institute for Occupational Health and Safety.
298. NIOSH. 1990. National Occupational Hazard Survey (1981-1983). National Institute for
Occupational Safety and Health. Last updated 7/1/90.
http://www.cdc.gov/noes/noes4/73470sco.html,
http://www.cdc.gov/noes/noes4/x8305sco.html.
299. NIOSH. 1994. Elements in Blood or Tissue: Method 8005. In NIOSH Manual of Analytical
Methods. 4th edition. National Institute for Occupational Health and Safety. 6 pp.
300. Nishioka H. 1975. Mutagenic activities of metal compounds in bacteria. Mutat Res 31(3):
185-189. (as cited in IARC 2006)
301. NIST. 2005. Why Are There No Volume Li-ion Battery Manufacturers in the United States:
Chapter II - Rationale for Li-ion Case National Institute of Standards and Technology.
Updated on 1/3/2007. http://www.atp.nist.gov/eao/wp05-01/chapt2.htm. Accessed on
5/22/15.
302. Nowak HF. 1966. Neoplasia in mouse skeletal muscles under the influence of polyester
resin activator. Arch Immunol Ther Exp (Warsz) 14(6): 774-778.
303. NTP. 1998. Toxicology and Carcinogenesis Studies of Cobalt Sulfate Heptahydrate (CAS
No. 10026-24-1) in F344/N Rats and B6C3F1/N Mice (Inhalation Studies). Technical
Report Series No. 471. NIH Publication No. 98-3961. Research Triangle Park, NC:
National Toxicology Program. 268 pp.
304. NTP. 2002b. Report on Carcinogens Background Document for Cobalt Sulfate. Research
Triangle Park, NC: National Toxicology Program. 71 pp.
305. NTP. 2009. Report on Carcinogens Background Document for Cobalt-Tungsten Carbide:
Powders and Hard metals. Research Triangle Park, NC: National Toxicology Program.
180 pp.
306. NTP. 2014a. Cobalt-tungsten carbide: Powders and hard metals. In Report on Carcinogens.
13th Edition. Research Triangle Park, NC: National Toxicology Program.
http://ntp.niehs.nih.gov/pubhealth/roc/roc13/index.html.
307. NTP. 2014b. Toxicology Studies of Cobalt Metal (CAS No. 7440-48-4) in F344/N Rats and
B6C3F1/N Mice and Toxicology and Carcinogenesis Studies of Cobalt Metal in
F344/NTac Rats and B6C3F1/N Mice (Inhalation Studies). Technical Report Series No.
166
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
581. NIH Publication No. 14-5932. Research Triangle Park, NC: National Toxicology
Program. 308 pp.
308. NTP. 2014c. Cobalt and Certain Cobalt Compounds Protocol. Research Triangle Park,
NC: National Toxicology Program. 50 pp.
http://ntp.niehs.nih.gov/ntp/roc/protocols/cobalt_508.pdf.
309. NTP. 2014d. Cobalt sulfate. In Report on Carcinogens. 13th Edition. Research Triangle
Park, NC: National Toxicology Program.
http://ntp.niehs.nih.gov/pubhealth/roc/roc13/index.html.
310. Nyga A, Hart A, Tetley TD. 2015. Importance of the HIF pathway in cobalt nanoparticleinduced cytotoxicity and inflammation in human macrophages. Nanotoxicology: 1-13.
(Supported by the Technology Strategy Board. Authors affiliated with Imperial College
London, UK; University College London, UK.)
311. O'Hara GP, Mann DE, Jr., Gautieri RF. 1971. Effect of cobalt chloride and sodium
cobaltinitrite on the growth of established epithelial tumors induced by
methylcholanthrene. J Pharm Sci 60(3): 473-474. (Support not reported. Authors affiliated
with American Foundation for Pharmaceutical Education.)
312. O'Rorke MA, Cantwell MM, Abnet CC, Brockman AJ, Murray LJ, Group FS. 2012.
Toenail trace element status and risk of Barrett's oesophagus and oesophageal
adenocarcinoma: results from the FINBAR study. Int J Cancer 131(8): 1882-1891.
(Supported by the Centre for Health Improvement, Queen’s University Belfast, Northern
Ireland, the Northern Ireland Research & Development Office, the Health Research Board,
Ireland, the Ulster Cancer Foundation and the Charitable Fund of the Royal Groups of
Hospitals, Belfast. Authors affiliated with Queens University Belfast, UK; NCI, MD;
University of Missouri-Columbia, MO.)
313. OECD. 2014. Environment Directorate Joint Meeting of the Chemicals Committee and the
Working Party On Chemicals, Pesticides and Biotechnology. Guidance on Grouping of
Chemicals, Second Edition. OECD Environment, Health and Safety Publications Series on
Testing and Assessment No. 194, ENV/JM/MONO(2014)4. Paris, France: Organisation for
Economic Co-operation and Development. 141 pp.
314. Ogawa HI, Sakata K, Inouye T, Jyosui S, Niyitani Y, Kakimoto K, Morishita M, Tsuruta S,
Kato Y. 1986. Combined mutagenicity of cobalt(II) salt and heteroaromatic compounds in
Salmonella typhimurium. Mutat Res 172(2): 97-104. (as cited in IARC 2006)
315. Ogawa HI, Shibahara T, Iwata H, Okada T, Tsuruta S, Kakimoto K, Sakata K, Kato Y, Ryo
H, Itoh T, et al. 1994. Genotoxic activities in vivo of cobaltous chloride and other metal
chlorides as assayed in the Drosophila wing spot test. Mutat Res 320(1-2): 133-140. (as
cited in IARC 2006)
316. Ogawa HI, Ohyama Y, Ohsumi Y, Kakimoto K, Kato Y, Shirai Y, Nunoshiba T,
Yamamoto K. 1999. Cobaltous chloride-induced mutagenesis in the supF tRNA gene of
Escherichia coli. Mutagenesis 14(2): 249-253. (as cited in IARC 2006)
This draft document should not be construed to represent final NTP determination or policy
167
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
317. Ohno I. 2010. Electroless Despoition of Alloys. In Modern Electroplating. 5th edition.
Schlesinger M, Paunovic M, eds.: John Wiley & Sons. pp. 499-506.
318. Ortega R, Bresson C, Darolles C, Gautier C, Roudeau S, Perrin L, Janin M, Floriani M,
Aloin V, Carmona A, Malard V. 2014. Low-solubility particles and a Trojan-horse type
mechanism of toxicity: the case of cobalt oxide on human lung cells. Part Fibre Toxicol
11: 14. (Supported by the CEA DSV (France) and Electricité De France (EDF) company.
Authors affiliated with University of Bordeaux, France; CEA, DEN, DPC, SEARS,
Laboratoire de développement Analytique Nucléaire, France; CEA, DSV, IBEB, Lab
Biochim System Perturb, France; CEA, DEN, DPC, SEARS, Laboratoire d’Analyse en
Soutien aux Exploitants, France; IRSN, France; University of Nîmes, France.)
319. Orzechowski RF, Gautieri RF, Mann DE, Jr. 1964. Effect of sodium cobaltinitrite on the
minimal carcinogenic dose-50 (Mcd-50) of methylcholanthrene in albino mice. J Pharm
Sci 53: 388-391. (Supported by NCI, NIH. Authors affiliated with Temple University, PA.)
320. Pagano DA, Zeiger E. 1992. Conditions for detecting the mutagenicity of divalent metals in
Salmonella typhimurium. Environ Mol Mutagen 19(2): 139-146. (as cited in IARC 2006)
321. Palecek E, Brazdova M, Cernocka H, Vlk D, Brazda V, Vojtesek B. 1999. Effect of
transition metals on binding of p53 protein to supercoiled DNA and to consensus sequence
in DNA fragments. Oncogene 18(24): 3617-3625. (Supported by the Grant Agency of the
Czech Republic, Volkswagen Stiftung, the Grant Agency of the Academy of Sciences CR
and IGA MH CR. Authors affiliated with Academy of Sciences of the Czech Republic,
Czech Republic; Masaryk Memorial Cancer Institute, Czech Repblic.)
322. Palit S, Sharma A, Talukder G. 1991a. Cytotoxic effects of cobalt chloride on mouse bone
marrow cells in vivo. Cytobios 68(273): 85-89. (Supported by the University Grants
Commission and Council of Scientific and Industrial Research. Authors affiliated with
University of Calcutta, India; Vivekananda Institute of Medical Sciences, India.)
323. Palit S, Ghosh AK, Sharma A, Talukder G. 1991b. Modification of the clastogenic effects
of cobalt by calcium in bone marrow cells of mice in vivo. Cytologia 56: 373-377. (As
cited in WHO 2006)
324. Palit S, Sharma A, Talukder G. 1991c. Chromosomal aberrations induced by cobaltous
chloride in mice in vivo. Biol Trace Elem Res 29(2): 139-145. (as cited in WHO 2006)
325. Palit S, Sharma A, Talukder G. 1991d. Protection by chlorophyllin against induction of
chromosomal aberrations by cobalt in bone marrow cells of mice in vivo. Fitoterapia
62(5): 425-428. (as cited in WHO 2006)
326. Papis E, Rossi F, Raspanti M, Dalle-Donne I, Colombo G, Milzani A, Bernardini G,
Gornati R. 2009. Engineered cobalt oxide nanoparticles readily enter cells. Toxicol Lett
189(3): 253-259. (Support not reported. Authors affiliated with Insubria University, Italy;
University of Milan, Italy; Centro di Ricerca Interuniversitario Politecnico di Milano e
Università dell’Insubria “The Protein Factory”, Italy.)
168
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
327. Park HS, Shin KO, Kim JS. 2007. Assessment of reference values for hair minerals of
Korean preschool children. Biol Trace Elem Res 116(2): 119-130.
328. Pasha Q, Malik SA, Iqbal J, Shah MH. 2007. Characterization and distribution of the
selected metals in the scalp hair of cancer patients in comparison with normal donors.
Biological Trace Element Research 118(3): 207-216. (Supported by the Higher Education
Commission, Government of Pakistan. Authors affiliated with Quaid-i-Azam University,
Pakistan; POF Hospital, Pakistan.)
329. Patel E, Lynch C, Ruff V, Reynolds M. 2012. Co-exposure to nickel and cobalt chloride
enhances cytotoxicity and oxidative stress in human lung epithelial cells. Toxicol Appl
Pharmacol 258(3): 367-375. (Support not reported. Authors affiliated with Washington
College, MD; U.S. Army Research Institute of Environmental Medicine, MA.)
330. Paton GR, Allison AC. 1972. Chromosome damage in human cell cultures induced by
metal salts. Mutat Res 16(3): 332-336. (as cited in IARC 2006)
331. Patrick G, Batchelor AL, Stirling C. 1989. An interspecies comparison of the lung
clearance of inhaled monodisperse cobalt oxide particles - Part VI: Lung clearance of
inhaled cobalt oxide particles in SPF Fischer rats. J Aerosol Sci 20(2): 249-255. (Support
not reported. Authors affiliated with MRC Radiobiology Unit, UK.)
332. Patrick G, Stirling C, Kreyling WG, Poncy JL, Duserre C, Collier CG, Godleski J, Brain
JD. 1994. Interspecies comparison of the clearance of ionic cobalt from the lngs. Inhal
Toxicol 6: 225-240. (Supported by the Commission of the European Communities. Authors
affiated with Medical Research Council Radiobiology Unit, UK; GSF-Forschungszentrcrm
für Umwelt und Gesundheit GmbH, Germany; Commissariat à I'Energie Atomique,
France; National Radiological Protection Board, UK; Harvard School of Public Health,
MA.)
333. Paul SAM, Simons JW, Mabjeesh NJ. 2004. HIF at the crossroads between ischemia and
carcinogenesis. Journal of Cellular Physiology 200(1): 20-30. (Supported by the Avon
Foundation, NIH, CaP CURE Foundation and the M.K. Humanitarian Association and
Parisian. Authors affiliated with Ludwig-Maximilians-University, Germany; Emory
University School of Medicine, GA; Tel Aviv Sourasky Medical Center, Israel.)
334. Pauluhn J. 2009. Comparative pulmonary response to inhaled nanostructures:
Considerations on test design and endpoints. Inhalation Toxicology 21(SUPPL. 1): 40-54.
(Supported by the German Federal Ministry of Education and Research. Author affiliated
with Bayer HealthCare AG, Germany.)
335. Paustenbach DJ, Tvermoes BE, Unice KM, Finley BL, Kerger BD. 2013. A review of the
health hazards posed by cobalt. Crit Rev Toxicol 43(4): 316-362. (Supported by DePuy
Orthopaedics, Inc. Authors affiliated with ChemRisk, LLC, CA, CO and PA.)
336. Pellet F, Perdrix A, Vincent M, Mallion J-M. 1984. [Biological levels of urinary cobalt].
Arch Mal Prof 45: 81-85. (as cited in IARC 2006 and IARC 1991)
This draft document should not be construed to represent final NTP determination or policy
169
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
337. Permenter MG, Dennis WE, Sutto TE, Jackson DA, Lewis JA, Stallings JD. 2013.
Exposure to cobalt causes transcriptomic and proteomic changes in two rat liver derived
cell lines. PLoS One 8(12): e83751. (Supported by the U.S. Army Medical Research and
Materiel Command, Military Operational Medicine Research Program. Authors affiliated
with Excet, Inc., MD; US Army Center for Environmental Health Research, MD; Naval
Research Laboratory, Washington, D.C.)
338. Petering HG, Yeager DW, Witherup SO. 1973. Trace metal content of hair. II. Cadmium
and lead of human hair in relation to age and sex. Arch Environ Health 27(5): 327-330.
(Supported by NIH. Authors affiliated with University of Cincinnati, OH.)
339. Peters K, Unger RE, Gatti AM, Sabbioni E, Tsaryk R, Kirkpatrick CJ. 2007. Metallic
nanoparticles exhibit paradoxical effects on oxidative stress and pro-inflammatory response
in endothelial cells in vitro. Int J Immunopathol Pharmacol 20(4): 679-689. (Supported by
the Deutsche Forschungsgemeinschaft and the European Commission. Authors affiliated
with Johannes Gutenberg University, Germany; University of Modena and Reggio Emilia,
Italy; European Commission, Italy.)
340. Petit A, Mwale F, Tkaczyk C, Antoniou J, Zukor DJ, Huk OL. 2005. Induction of protein
oxidation by cobalt and chromium ions in human U937 macrophages. Biomaterials 26(21):
4416-4422. (Supported by the Orthopaedic Research Fund of the Sir Mortimer B. Davis–
Jewish General Hospital. Authors affiliated with McGill University, Canada.)
341. Pihl RO, Parkes M. 1977. Hair element content in learning disabled children. Science
198(4313): 204-206.
342. Polyzois I, Nikolopoulos D, Michos I, Patsouris E, Theocharis S. 2012. Local and systemic
toxicity of nanoscale debris particles in total hip arthroplasty. J Appl Toxicol 32(4): 255269. (Support not reported. Authors affiliated with National and Kapodistrian University of
Athens, Greece; Ascleipion Voulas General Hospital, Greece.)
343. Ponti J, Sabbioni E, Munaro B, Broggi F, Marmorato P, Franchini F, Colognato R, Rossi F.
2009. Genotoxicity and morphological transformation induced by cobalt nanoparticles and
cobalt chloride: an in vitro study in Balb/3T3 mouse fibroblasts. Mutagenesis 24(5): 439445. (Supported by the European Commission, Joint Research Centre, Nanobiosciences
Unit, project: ‘NanoBioTechnology for Health’. Authors affiliated with European
Commission, Italy.)
344. Posma FD, Dijstelberger SK. 1985. Serum and urinary cobalt levels as indicators of cobalt
exposure in hard metal workers. In Proceedings of an International Conference, Heavy
Metals in the Environment, Athens, September 1985. Lekkas TD, ed. Luxembourg:
Commission of the European Communities. pp. 89-91. (as cited in IARC 2006 and IARC
1991)
345. Poulsen OM, Olsen E, Christensen JM, Vinzent P, Petersen OH. 1995. Geltape method for
measurement of work related surface contamination with cobalt containing dust:
Correlation between surface contamination and airborne exposure. Occup Environ Med 52:
827-833. (as cited in IARC 2006)
170
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
346. Pourahmad J, O'Brien PJ, Jokar F, Daraei B. 2003. Carcinogenic metal induced sites of
reactive oxygen species formation in hepatocytes. Toxicol In Vitro 17(5-6): 803-810.
(Support not reported. Authors affiliated with Shaheed Beheshti University of Medical
Sciences, Iran; University of Toronto, Canada.)
347. Prazmo W, Balbin E, Baranowska H, Ejchart A, Putrament A. 1975. Manganese
mutagenesis in yeast. II. Conditions of induction and characteristics of mitochondrial
respiratory deficient Saccharomyces cerevisiae mutants induced with manganese and
cobalt. Genet Res 26(1): 21-29. (as cited in IARC 2006)
348. Prescott E, Netterstrom B, Faber J, Hegedus L, Suadicani P, Christensen JM. 1992. Effect
of occupational exposure to cobalt blue dyes on the thyroid volume and function of female
plate painters. Scand J Work Environ Health 18(2): 101-104. (Supported by SID and Royal
Copenhagen. Authors affiliated with University Hospital of Copenhagen, Denmark; Herlev
University Hospi, Denmark; Frederiksb erg Hospital, Denmark; Danish National Institute
of Occupational Health, Denmark.)
349. PubChem. 2015. PubChem Compound. National Center for Biotechnology Information.
http://www.ncbi.nlm.nih.gov/pccompound and search by Compound Identification
Number. Accessed on 3/24/15.
350. Pulido MD, Parrish AR. 2003. Metal-induced apoptosis: mechanisms. Mutat Res 533(1-2):
227-241. (Support not reported. Authors affiliated with Texas A&M University System
Health Science Center, TX.)
351. Putrament A, Baranowska H, Ejchart A, Jachymczyk W. 1977. Manganese mutagenesis in
yeast. VI. Mn2+ uptake, mitDNA replication and ER induction: comparison with other
divalent cations. Mol Gen Genet 151(1): 69-76. (as cited in IARC 2006)
352. Qayyum MA, Shah MH. 2014. Comparative Assessment of Selected Metals in the Scalp
Hair and Nails of Lung Cancer Patients and Controls. Biological Trace Element Research
158(3): 305-322. (Supported by the Higher Education Commission, Government of
Pakistan. Authors affiliated with Quaid-i-Azam University, Pakistan.)
353. Qiao H, Li L, Qu ZC, May JM. 2009. Cobalt-induced oxidant stress in cultured endothelial
cells: prevention by ascorbate in relation to HIF-1alpha. Biofactors 35(3): 306-313.
(Supported by NIH and the Vanderbilt Diabetes Research and Training Center. Authors
affiliated with Vanderbilt University School of Medicine, TN.)
354. Qiao Y, Ma L. 2013. Quantification of metal ion induced DNA damage with single cell
array based assay. Analyst 138(19): 5713-5718. (Supported by Bankhead-Copley Cancer
Research Program of Florida Department of Health. Authors affiliated with University of
Central Florida, FL.)
355. Raabe GK, Collingwood KW, Wong O. 1998. An updated mortality study of workers at a
petroleum refinery in Beaumont, Texas. Am J Ind Med 33(1): 61-81. (Support not reported.
Authors affiliated with Mobil Business Resources Corporation, PA; Applied Health
Sciences, Inc., CA.)
This draft document should not be construed to represent final NTP determination or policy
171
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
356. Raffn E, Mikkelsen S, Altman DG, Christensen JM, Groth S. 1988. Health effects due to
occupational exposure to cobalt blue dye among plate painters in a porcelain factory in
Denmark. Scand J Work Environ Health 14(6): 378-384. (Supported by the Danish Work
Environment Fund. Authors affiliated with Danish Labour Inspectorate, Denmark;
Rigshospitale, Denmark; Panum Institute, Denmark; Clinical Research Centre UK; Danish
National Institute of Occupational Health, Denmark; Finsen Institute, Denmark.)
357. Rasgele PG, Kekecoglu M, Muranli FDG. 2013. Induction of micronuclei in mice bone
marrow cells by cobalt and copper chlorides. Arch Environ Prot 39(1): 75-82. (Support not
reported. Authors affiliated with Duzce University, Turkey; Trakya University, Turkey.)
358. Reddy SB, Charles MJ, Kumar MR, Reddy BS, Anjaneyulu C, Raju GJN, Sundareswar B,
Vijayan V. 2002. Trace elemental analysis of adenoma and carcinoma thyroid by PIXE
method. Nucl Instr Methods Phys Res B 196: 333-339.
359. Reinardy HC, Syrett JR, Jeffree RA, Henry TB, Jha AN. 2013. Cobalt-induced
genotoxicity in male zebrafish (Danio rerio), with implications for reproduction and
expression of DNA repair genes. Aquat Toxicol 126: 224-230. (Supported by the
Government of the Principality of Monaco, the International Atomic Energy Agency and
Plymouth University Marine Institute. Authors affiliated with University of Plymouth, UK;
University of Technology, Australia; University of Tennessee, TN.)
360. Resende de Souza Nazareth H. 1976. [Efeito do cloreto de cobalto em não-disjunção]. Cie
Cult 28: 1472-1475. (as cited in IARC 2006)
361. Richardson HW, Meshri DT. 2001. Cobalt compounds. In Kirk-Othmer Encyclopedia of
Chemical Technology. vol. 7. John Wiley & Sons, Inc. pp. 229-249.
362. Richter PA, Bishop EE, Wang J, Swahn MH. 2009. Tobacco smoke exposure and levels of
urinary metals in the U.S. youth and adult population: the National Health and Nutrition
Examination Survey (NHANES) 1999-2004. Int J Environ Res Public Health 6(7): 19301946. (Support not reported. Authors affiliated with CDC, GA; RTI International, GA;
Georgia State University, GA.)
363. Robison SH, Cantoni O, Costa M. 1982. Strand breakage and decreased molecular weight
of DNA induced by specific metal compounds. Carcinogenesis 3(6): 657-662. (as cited in
IARC 2006)
364. Rodriguez de la Flor M, Hernandez-Vaquero D, Fernandez-Carreira JM. 2013. Metal
presence in hair after metal-on-metal resurfacing arthroplasty. J Orthop Res 31(12): 20252031.
365. Rogers MA, Thomas DB, Davis S, Vaughan TL, Nevissi AE. 1993. A case-control study of
element levels and cancer of the upper aerodigestive tract. Cancer Epidemiol Biomarkers
Prev 2(4): 305-312. (Supported by NCI. Authors affiliated with SUNY Health Science
Center, NY; Fred Hutchinson Cancer Research Center, WA; University of Washington,
WA.)
172
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
366. Romero A, Ramos E, De Los Ríos C, Egea J, Del Pino J, Reiter RJ. 2014. A review of
metal-catalyzed molecular damage: Protection by melatonin. Journal of Pineal Research
56(4): 343-370. (Supported by the Miguel Servet Program. Authors affiliated with
Universidad Complutense de Madrid, Spain; Universidad Autonoma de Madrid, Spain;
Hospital Universitario de la Princesa, Spain; University of Texas Health Science, Center at
San Antonio, TX.)
367. Rossman TG, Molina M, Meyer LW. 1984. The genetic toxicology of metal compounds: I.
Induction of lambda prophage in E coli WP2s(lambda). Environ Mutagen 6(1): 59-69. (as
cited in IARC 2006)
368. Rudge CV, Röllin HB, Nogueira CM, Thomassen Y, Rudge MC, Odland JØ. 2009. The
placenta as a barrier for toxic and essential elements in paired maternal and cord blood
samples of South African delivering women. J Environ Monit 11(7): 1322-1330.
(Supported by the Brazilian Federal Agency for Graduate Studies. Authors affiliated with
São Paulo State University, Brazil; Medical Research Council, South Africa; University of
Pretoria, South Africa; University of the Witwatersrand, South Africa; National Institute
for Occupational Health, South Africa; National Institute for Occupational Health, Norway;
University of Tromsø, Norway; University of Aarhus, Denmark.)
369. Rufe PF. 2010. Testimony for United States Senate Committee on Energy and Natural
Resources, September 30, 2010. Salmon, ID: Formation Capital Corporation, U.S. 164 pp.
370. Saat NZM, Chow SY, Ghazali AR, Hamid ZA, Lubis SH, Mohamed N, Ishak I, Othman H.
2013. Study of heavy metal levels in nails and hairs among vegetable farmers in Malaysia.
Res J Appl Sci 8(9): 449-455.
371. Sabbioni E, Minoia C, Pietra R, Mosconi G, Forni A, Scansetti G. 1994a. Metal
determinations in biological specimens of diseased and non-diseased hard metal workers.
Sci Total Environ 150(1-3): 41-54. (Support not reported. Authors affiliated with European
Commission, Italy; Fondazione Clinica del Lavoro, Italy; Ospedali Riuniti di Bergamo,
Italy; University of Milan, Italy; University of Turin, Italy.)
372. Sabbioni E, Fortaner S, Farina M, Del Torchio R, Petrarca C, Bernardini G, MarianiCostantini R, Perconti S, Di Giampaolo L, Gornati R, Di Gioacchino M. 2014. Interaction
with culture medium components, cellular uptake and intracellular distribution of cobalt
nanoparticles, microparticles and ions in Balb/3T3 mouse fibroblasts. Nanotoxicology 8(1):
88-99. (Supported by the Unit of “Immunotoxicology and Allergy” of Ce.S.I., ECVAM
Unit, the “Fondi Ateneo per la Ricerca” from the University of Insubria and the
“Associazione Italiana Ricerca sul Cancro." Authors affiliated with European Center for
the Sustainable Impact of Nanotechnologies, Italy; European Commission, Italy; PhiScience srl Chemical and Safety Consulting, Italy; Ext-EC-Europe, Italy; Aging Research
Center, Italy; University of Insubria, Italy; Polytechnic University of Milan, Italy;
d’Annunzio University, Italy.)
373. Saini Y, Greenwood KK, Merrill C, Kim KY, Patial S, Parameswaran N, Harkema JR,
LaPres JJ. 2010a. Acute cobalt-induced lung injury and the role of hypoxia-inducible factor
This draft document should not be construed to represent final NTP determination or policy
173
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
1 alpha in modulating inflammation. Toxicological Sciences 116(2): 673-681. (Supported
by NIH. Authors affiliated with Michigan State University, MI.)
374. Saini Y, Kim KY, Lewandowski R, Bramble LA, Harkema JR, Lapres JJ. 2010b. Role of
hypoxia-inducible factor 1{alpha} in modulating cobalt-induced lung inflammation. Am J
Physiol Lung Cell Mol Physiol 298(2): L139-147. (Supported by NIEHS. Authors affiliated
with Michigan State University, MI.)
375. Salnikow K, Su W, Blagosklonny MV, Costa M. 2000. Carcinogenic metals induce
hypoxia-inducible factor-stimulated transcription by reactive oxygen species-independent
mechanism. Cancer Res 60(13): 3375-3378. (Supported by NIH. Authors affiliated with
New York University School of Medicine, NY; NIH, MD.)
376. Salnikow K, Donald SP, Bruick RK, Zhitkovich A, Phang JM, Kasprzak KS. 2004.
Depletion of intracellular ascorbate by the carcinogenic metals nickel and cobalt results in
the induction of hypoxic stress. J Biol Chem 279(39): 40337-40344. (Supported by NIEHS.
Authors affiliated with NIH, MD; University of Texas Southwestern Medical Center, TX;
Brown University, RI.)
377. Sampson B, Hart A. 2012. Clinical usefulness of blood metal measurements to assess the
failure of metal-on-metal hip implants. Ann Clin Biochem 49(Pt 2): 118-131. (Supported by
the The London Implant Retrieval Centre (LIRC), which is funded by the British
Orthopaedic Association (BOA). Authors affiliated with Imperial College Healthcare NHS
Trust, UK; Imperial College, UK.)
378. Scansetti G, Lamon S, Talarico S, Botta GC, Spinelli P, Sulotto F, Fantoni F. 1985. Urinary
cobalt as a measure of exposure in the hard metal industry. Int Arch Occup Environ Health
57(1): 19-26. (as cited in IARC 2006 and IARC 1991)
379. Scansetti G, Botta GC, Spinelli P, Reviglione L, Ponzetti C. 1994. Absorption and
excretion of cobalt in the hard metal industry. Sci Total Environ 150(1-3): 141-144.
(Support not reported. Authors affiliated with Universita di Torino, Italy; Servizio di Igiene
Pubblica, Italy.)
380. Scansetti G, Maina G, Botta GC, Bambace P, Spinelli P. 1998. Exposure to cobalt and
nickel in the hard-metal production industry. Int Arch Occup Environ Health 71(1): 60-63.
(Support not reported. Authors affiliated with Università degli Studi di Torino, Italy.)
381. Schade SG, Felsher BF, Bernier GM, Conrad ME. 1970. Interrelationship of cobalt and
iron absorption. J Lab Clin Med 75(3): 435-441. (Support not reported. Authors affiliated
with Walter Reed Army Institute of Research, Washington, D.C.)
382. Schaffer AW, Pilger A, Engelhardt C, Zweymueller K, Ruediger HW. 1999. Increased
blood cobalt and chromium after total hip replacement. J Toxicol Clin Toxicol 37(7): 839844.
383. Scharf B, Clement CC, Zolla V, Perino G, Yan B, Elci SG, Purdue E, Goldring S,
Macaluso F, Cobelli N, Vachet RW, Santambrogio L. 2014. Molecular analysis of
chromium and cobalt-related toxicity. Sci Rep 4: 5729.
174
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
384. Schrauzer GN, Shrestha KP, Flores-Arce MF. 1992. Lithium in scalp hair of adults,
students, and violent criminals. Effects of supplementation and evidence for interactions of
lithium with vitamin B12 and with other trace elements. Biol Trace Elem Res 34(2): 161176.
385. Schumacher-Wittkopf E, Angerer J. 1981. [A practical method for the determination of
cobalt in urine (author's transl)]. Int Arch Occup Environ Health 49(1): 77-81. (as cited in
IARC 1991)
386. SciFinder. 2015. SciFinder. http://www.cas.org/products/scifinder. Accessed on 4/30/15.
387. SEER. 2015a. SEER Cancer Statistics Factsheets: Lung and Bronchus Cancer. National
Cancer Institute. http://seer.cancer.gov/statfacts/html/lungb.html. Accessed on 3/30/15.
388. SEER. 2015b. SEER Cancer Statistics Factsheets: Espohageal Cancer. National Cancer
Institute. http://seer.cancer.gov/statfacts/html/esoph.html. Accessed on 3/30/15.
389. SEER. 2015c. SEER Cancer Stat Fact Sheets. National Cancer Institute.
http://seer.cancer.gov/statfacts/. Accessed on 3/30/15.
390. Shabaan AA, Marks V, Lancaster MC, Dufeu GN. 1977. Fibrosarcomas induced by cobalt
chloride (CoCl2) in rats. Lab Anim 11(1): 43-46.
391. Shacklette HT, Boerngen JG. 1984. Element concentrations in soils and other surficial
materials of the conterminous United States. U.S. Geological Survey Professional Paper
1270. Washington, D.C.
392. Sharma P. 2004. Review article: prevalence of Barrett's oesophagus and metaplasia at the
gastro-oesophageal junction. Aliment Pharmacol Ther 20 Suppl 5: 48-54; discussion 61-42.
393. Shedd KB. 1993. The Materials Flow of Cobalt in the United States. IC 9350. United
States Department of the Interior. 31 pp.
394. Shedd KB. 2014a. U.S. Geological Survey, Mineral Commodity Summaries, February
2014: Cobalt. http://minerals.usgs.gov/minerals/pubs/commodity/cobalt/. Accessed on
2/12/15.
395. Shedd KB. 2014b. USGS 2012 Minerals Yearbook: Cobalt [Advance Release].
http://minerals.usgs.gov/minerals/pubs/commodity/cobalt/index.html - myb. Accessed on
2/12/15.
396. Shi X, Dalal NS, Kasprzak KS. 1993. Generation of free radicals from model lipid
hydroperoxides and H2O2 by Co(II) in the presence of cysteinyl and histidyl chelators.
Chem Res Toxicol 6(3): 277-283.
397. Shukla SJ, Huang R, Simmons SO, Tice RR, Witt KL, Vanleer D, Ramabhadran R, Austin
CP, Xia M. 2012. Profiling environmental chemicals for activity in the antioxidant
response element signaling pathway using a high throughput screening approach. Environ
Health Perspect 120(8): 1150-1156.
This draft document should not be construed to represent final NTP determination or policy
175
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
398. Sighinolfi GL, Artoni E, Gatti AM, Corsi L. 2014. Carcinogenic potential of metal
nanoparticles in BALB/3T3 cell transformation assay. Environ Toxicol. (Support not
reported. Authors affiliated with University of Modena and Reggio Emilia, Italy; Institute
for Advanced Sciences Convergence & Int’l Clean Water Institute, VA.)
399. Simcox NJ, Stebbins A, Guffey S, Atallah R, Hibbard R, Camp J. 2000. Hard metal
exposures. Part 2: Prospective exposure assessment. Appl Occup Environ Hyg 15(4): 342353. (Support not reported. Authors affiliated with University of Washington, WA.)
400. Simonsen LO, Brown AM, Harbak H, Kristensen BI, Bennekou P. 2011. Cobalt uptake and
binding in human red blood cells. Blood Cells Mol Dis 46(4): 266-276. (Support not
reported. Authors affiliated with University of Copenhagen, Denmark; University of
Nottingham, UK.)
401. Simonsen LO, Harbak H, Bennekou P. 2012. Cobalt metabolism and toxicology--a brief
update. Sci Total Environ 432: 210-215. (Supoprt not reported. Authors affiliated with
University of Copenhagen, Denmark.)
402. Singh I. 1983. Induction of reverse mutation and mitotic gene conversion by some metal
compounds in Saccharomyces cerevisiae. Mutat Res 117(1-2): 149-152. (as cited in IARC
2006)
403. Slotnick MJ, Nriagu JO. 2006. Validity of human nails as a biomarker of arsenic and
selenium exposure: A review. Environ Res 102(1): 125-139.
404. Smith IC, Carson BL. 1981. Section C. Geochemistry and Occurence. Section D.
Environmental Transport. In Trace Metals in the Environment. vol. 6 - Cobalt. Smith IC,
Carson BL, eds. Ann Arbor, MI: Ann Arbor Science Publishers. pp. 359-367. (as cited in
Paustenbach et al. 2013)
405. Smith LJ, Holmes AL, Kandpal SK, Mason MD, Zheng T, Wise JP, Sr. 2014. The
cytotoxicity and genotoxicity of soluble and particulate cobalt in human lung fibroblast
cells. Toxicol Appl Pharmacol 278(3): 259-265. (Supported by ARO and the Maine Center
for Toxicology and Environmental Health. Authors afiliated with University of Southern
Maine, ME; University of Maine, ME; Yale School of Public Health, CT.)
406. Smith T, Edmonds CJ, Barnaby CF. 1972. Absorption and retention of cobalt in man by
whole-body counting. Health Phys 22(4): 359-367. (Support not reported. Authors
affiliated with University College Hospital Medical School, UK.)
407. Snyder RD, Davis GF, Lachmann PJ. 1989. Inhibition by metals of X-ray and ultravioletinduced DNA repair in human cells. Biol Trace Elem Res 21: 389-398. (as cited in IARC
2006)
408. Sorbie J, Olatunbosun D, Corbett WE, Valberg LS. 1971. Cobalt excretion test for the
assessment of body iron stores. Can Med Assoc J 104(9): 777-782. (Support not reported.
Authors affiliated with Queen's University, Canada; Kingston General Hospital, Canada.)
409. Sprince NL, Chamberlin RI, Hales CA, Weber AL, Kazemi H. 1984. Respiratory disease in
tungsten carbide production workers. Chest 86(4): 549-557. (Support not reported. Authors
176
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
affiliated with Harvard Medical School, MA; Massachusetts Institute of Technology, MA;
Massachusetts Eye and Ear Infirmary, MA.)
410. Steinhoff D, Mohr U. 1991. On the question of a carcinogenic action of cobalt-containing
compounds. Exp Pathol 41(4): 169-174.
411. Stopford W, Turner J, Cappellini D, Brock T. 2003. Bioaccessibility testing of cobalt
compounds. J Environ Monit 5(4): 675-680.
412. Suardi R, Belotti L, Ferrari MT, Leghissa P, Caironi M, Maggi L, Alborghetti F, Storto T,
Silva T, Piazzolla S. 1994. Health survey of workers occupationally exposed to cobalt. Sci
Total Environ 150(1-3): 197-200. (Support not reported. Authors affiliated with Local
Sanitary Unit 29, Italy; Ospedali Riuniti of Bergamo, Italy; Local Sanitary Unit 28, Italy;
Local Sanitary Unit 30, Italy.)
413. Sunderman FW, Jr., Hopfer SM, Swift T, Rezuke WN, Ziebka L, Highman P, Edwards B,
Folcik M, Gossling HR. 1989b. Cobalt, chromium, and nickel concentrations in body fluids
of patients with porous-coated knee or hip prostheses. J Orthop Res 7(3): 307-315.
(Supported by Howmedica Inc. and the Northeast Utilities Company. Authors affiliated
with University of Connecticut School of Medicine, CT.)
414. Suzuki T, Yamamoto R. 1982. Organic mercury levels in human hair with and without
storage for eleven years. Bull Environ Contam Toxicol 28(2): 186-188. (Supported by the
Japanese Ministry of Education and Culture. Authors affiliated with University of Tokyo,
Japan; Tohoku University School of Medicine, Japan.)
415. Suzuki Y, Shimizu H, Nagae Y, Fukumoto M, Okonogi H, Kadokura M. 1993.
Micronucleus test and erythropoiesis: effect of cobalt on the induction of micronuclei by
mutagens. Environ Mol Mutagen 22(2): 101-106. (Support not reported. Authors affiliated
with Jikei University School of Medicine, Japan; CIBA-GEIGY, Japan; Takarazuka,
Japan.)
416. Swennen B, Buchet JP, Stanescu D, Lison D, Lauwerys R. 1993. Epidemiological survey
of workers exposed to cobalt oxides, cobalt salts, and cobalt metal. Br J Ind Med 50(9):
835-842.
417. Talakin YN, Ivanova LA, Kostetskaya NI, Komissarov VN, Belyaeva IV. 1991. [Hygienic
characteristics of working conditions and health state of workers engaged in the production
of cobalt salts]. Gig Tr Prof Zbl 1(10-11). (as cited in IARC 2006)
418. Thomas RG, Furchner JE, London JE, Drake GA, Wilson JS, Richmond CR. 1976.
Comparative metabolism of radionuclides in mammals--X. Retention of tracer-level cobalt
in the mouse, rat, monkey and dog. Health Phys 31(4): 323-333. (Supported by the U.S.
Energy Research and Development Administration. Authors affiliated with Holifield
National Laboratory, TN; Los Alamos Scientific Laboratory, NM.)
419. Thomassen Y, Nieboer E, Ellingsen D, Hetland S, Norseth T, Odland JO, Romanova N,
Chernova S, Tchachtchine VP. 1999. Characterisation of workers' exposure in a Russian
nickel refinery. J Environ Monit 1(1): 15-22.
This draft document should not be construed to represent final NTP determination or policy
177
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
420. Thompson RS, Gautieri RF, Mann DE, Jr. 1965. Effect of chronic oral administration of
sodium cobaltinitrite and sodium nitrite on the minimal carcinogenic dose50 of
methylcholanthrene in albino mice. J Pharm Sci 54(4): 595-598.
421. Thomson AB, Valberg LS, Sinclair DG. 1971. Competitive nature of the intestinal
transport mechanism for cobalt and iron in the rat. J Clin Invest 50(11): 2384-2394.
(Supported by the Medical Research Council. Authors affiliated with Queens University,
Canada.)
422. TRI. 2014a. TRI Explorer Chemical Report. On-site Disposal to Class I Underground
Injection Wells, RCRA Subtitle C Landfills, and Other Landfills. U.S. Environmental
Protection Agency. http://www.epa.gov/triexplorer. Last accessed on 10/13/14.
423. TRI. 2014b. TRI Explorer Chemical Report. TRI On-site and Off-site Reported Disposed of
or Otherwise Released (in pounds), Trend Report for Facilities in All industries, for Cobalt
Chemical, U.S. U.S. Environmental Protection Agency. http://www.epa.gov/triexplorer.
Last accessed on 10/14/14.
424. TRI. 2014c. TRI Explorer Chemical Report. TRI On-­site and Off-­site Reported Disposed
of or Otherwise Released (in pounds), Trend Report for Facilities in All industries, for
Cobalt Compounds Chemical, U.S. U.S. Environmental Protection Agency.
http://www.epa.gov/triexplorer. Last accessed on 10/13/14.
425. TRI. 2014d. TRI EZ Search in Envirofacts. Toxic Chemical Releases to the Environment
(in pounds), for Cobalt and Cobalt Compounds, U.S. U.S. Environmental Protection
Agency. http://www.epa.gov/enviro/facts/tri/ez.html. Last accessed on 10/29/14.
426. Tso WW, Fung WP. 1981. Mutagenicity of metallic cations. Toxicol Lett 8(4-5): 195-200.
(as cited in IARC 2006)
427. Tüchsen F, Jensen MV, Villadsen E, Lynge E. 1996. Incidence of lung cancer among
cobalt-exposed women. Scand J Work Environ Health 22(6): 444-450.
428. Tvermoes BE, Finley BL, Unice KM, Otani JM, Paustenbach DJ, Galbraith DA. 2013.
Cobalt whole blood concentrations in healthy adult male volunteers following two-weeks
of ingesting a cobalt supplement. Food Chem Toxicol 53: 432-439. (Supported by DePuy
Orthopaedics, Inc. Authors affiliated with ChemRisk, LLC, CO, CA and PA.)
429. Tvermoes BE, Unice KM, Paustenbach DJ, Finley BL, Otani JM, Galbraith DA. 2014.
Effects and blood concentrations of cobalt after ingestion of 1 mg/d by human volunteers
for 90 d. Am J Clin Nutr 99(3): 632-646.
430. Underwood EJ. 1977. Cobalt. In Trace Elements in Human and Animal Nutrition. 4th ed.
New York: Academic Press. p. 132-158.
431. Unice KM, Monnot AD, Gaffney SH, Tvermoes BE, Thuett KA, Paustenbach DJ, Finley
BL. 2012. Inorganic cobalt supplementation: prediction of cobalt levels in whole blood and
urine using a biokinetic model. Food Chem Toxicol 50(7): 2456-2461. (Supported by
DePuy Orthopaedics. Authors affiliated with ChemRisk, LLC, CA, PA and CO.)
178
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
432. Unice KM, Kerger BD, Paustenbach DJ, Finley BL, Tvermoes BE. 2014. Refined
biokinetic model for humans exposed to cobalt dietary supplements and other sources of
systemic cobalt exposure. Chem Biol Interact 216: 53-74.
433. US Steel. 2014. Sheet Products: Cold Rolled Sheet, Vitreous Enamel Sheet. United States
Steel Corporation. http://www.ussteel.com/uss/portal/home/products/sheet/Cold Rolled
Sheet/Products-Sheet-ColdRolledSheetVitreousEnamelSheet/!ut/p/b1/vZPLbsIwEEW_qMzYceJkGUhwTIht8gKyQbRUFe8KKL
T5elKp6qJSy6ZhZjXSlc7M1VyoYEo8y7WYZ1MOE6h28_PyZX5a7nfzzedcOTNRUMqi
hGtXlBylMbaTxilFyWAM0wimjYrPYmSiHLI0M9prVLkgsfYM9RmFHCbIZtnq41XW6
zpd4XucrRKSH8YXDInO6mGpijVVaqHy8EIQSZwXrE7QU8ljX55UYBZlWnR9PxxM6
s0tIN4ZOHJaAn4br4shQWlTE_gJQyFY20AXR2EDdHuECptoQ1q31IkClIESPQs5Nnvc2
VLkbV9Il3WXBhnQucuxd7_P80AqrfjsXN52nYoR8uDsWwCWv2ZYUpuCLT1JcBfykdQ0X77.
Accessed on 1/30/15.
434. USITC. 2014. USITC Interactive Tariff and Trade Dataweb. United States International
Trade Commission. http://dataweb.usitc.gov/scripts/user_set.asp and search on HTS no.
2836999050, 2836991000, 2833291000, 2833299100, 8105200000, 8105209000,
8105203000, 8105206000, 2836991000, 2833291000, 8105209000, 8105203000,
8105900000, 2915293000, 2827396000, 2605000000, 2822000000, 8105300000.
Accessed on 10/14.
435. Vaca CE, Wilhelm J, Harms-Ringdahl M. 1988. Interaction of lipid peroxidation products
with DNA. A review. Mutat Res 195(2): 137-149.
436. Valberg LS, Ludwig J, Olatunbosun D. 1969. Alteration in cobalt absorption in patients
with disorders of iron metabolism. Gastroenterology 56(2): 241-251. (Supported by the
Medical Research Council of Canada. Authors affiliated with Kingston General Hospital,
Canada; Queen's University, Canada.)
437. Vales G, Demir E, Kaya B, Creus A, Marcos R. 2013. Genotoxicity of cobalt nanoparticles
and ions in Drosophila. Nanotoxicology 7(4): 462-468. (Supported by the Universitat
Autònoma de Barcelona, the Akdeniz University and the Council of Higher Education, the
Generalitat de Catalunya and the the Management Unit of Research Projects of Akdeniz
University. Authors affiliated with Universitat Autònoma de Barcelona, Spain; Akdeniz
University, Turkey; CIBER Epidemiología y Salud Pública, Spain.)
438. Valko M, Morris H, Cronin MT. 2005. Metals, toxicity and oxidative stress. Curr Med
Chem 12(10): 1161-1208.
439. Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M. 2006. Free radicals, metals and
antioxidants in oxidative stress-induced cancer. Chem Biol Interact 160(1): 1-40.
440. van den Oever R, Roosels D, Douwen M, Vanderkeel J, Lahaye D. 1990. Exposure of
diamond polishers to cobalt. Ann Occup Hyg 34(6): 609-614. (Support not reported.
Authors affiliated with National Confederation of Christian Sickness Funds, Belgium; Fund
of Occupational Diseases, Belgium; Metallurgie Hoboken-Overpelt Sint Theresiastraat,
Belgium.)
This draft document should not be construed to represent final NTP determination or policy
179
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
441. Van Goethem F, Lison D, Kirsch-Volders M. 1997. Comparative evaluation of the in vitro
micronucleus test and the alkaline single cell gel electrophoresis assay for the detection of
DNA damaging agents: genotoxic effects of cobalt powder, tungsten carbide and cobalttungsten carbide. Mutat Res 392(1-2): 31-43. (Supported by the Belgian Federal Office for
Scientific, Technical and Cultural affairs. Authors affiliated with Free UniÍersity Brussels,
Belgium; Catholic UniÍersity of Louvain, Belgium.)
442. Vengellur A, Phillips JM, Hogenesch JB, LaPres JJ. 2005. Gene expression profiling of
hypoxia signaling in human hepatocellular carcinoma cells. Physiol Genomics 22(3): 308318. (Support not reported. Authors affiliated with Michigan State University, MI;
Genomics Institute of the Novartis Research Foundation, CA.)
443. Verstraelen S, Remy S, Casals E, De Boever P, Witters H, Gatti A, Puntes V, Nelissen I.
2014. Gene expression profiles reveal distinct immunological responses of cobalt and
cerium dioxide nanoparticles in two in vitro lung epithelial cell models. Toxicol Lett
228(3): 157-169.
444. Voroshilin SI, Plotko EG, Fink TV, Nikiforova V. 1978. [Cytogenetic effect of inorganic
wolfram, zinc, cadmium and cobalt compounds on human and animal somatic cells]. Tsitol
Genet 12(3): 241-243. (as cited in IARC 2006)
445. Wan R, Mo Y, Feng L, Chien S, Tollerud DJ, Zhang Q. 2012. DNA damage caused by
metal nanoparticles: involvement of oxidative stress and activation of ATM. Chem Res
Toxicol 25(7): 1402-1411.
446. Wang D, Tian X, Jiang Y. 2012. NDRG1/Cap43 overexpression in tumor tissues and serum
from lung cancer patients. J Cancer Res Clin Oncol 138: 1813-1820.
447. Wang G, Hazra TK, Mitra S, Lee HM, Englander EW. 2000. Mitochondrial DNA damage
and a hypoxic response are induced by CoCl(2) in rat neuronal PC12 cells. Nucleic Acids
Res 28(10): 2135-2140.
448. Wang GL, Semenza GL. 1995. Purification and characterization of hypoxia-inducible
factor 1. J Biol Chem 270(3): 1230-1237.
449. Wappelhorst O, Kuhn I, Heidenreich H, Markert B. 2002. Transfer of selected elements
from food into human milk. Nutrition 18(4): 316-322.
450. Webb M, Heath JC, Hopkins T. 1972. Intranuclear distribution of the inducing metal in
primary rhabdomyosarcomata induced in the rat by nickel, cobalt and cadmium. Br J
Cancer 26(4): 274-278.
451. Wedrychowski A, Schmidt WN, Hnilica LS. 1986. DNA-protein crosslinking by heavy
metals in Novikoff hepatoma. Arch Biochem Biophys 251(2): 397-402.
452. Wehner AP, Craig DK. 1972. Toxicology of inhaled NiO and CoO in Syrian golden
hamsters. Am Ind Hyg Assoc J 33(3): 146-155. (Supported by NCI. Authors affiliated with
Battelle Pacific Northwest Laboratories, WA.)
180
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
453. Wehner AP, Busch RH, Olson RJ, Craig DK. 1977. Chronic inhalation of cobalt oxide and
cigarette smoke by hamsters. Am Ind Hyg Assoc J 38(7): 338-346.
454. White MA, Dyne D. 1994. Biological monitoring of occupational cobalt exposure in the
United Kingdom. Sci Total Environ 150(1-3): 209-213. (Support not reported. Authors
affiliated with Health and Safety Executive, UK. )
455. WHO. 2006. Cobalt and Inorganic Cobalt Compounds. Concise International Chemical
Assessment Document 69. Geneva, Switzerland: World Health Organization. 93 pp.
456. Wild P, Perdrix A, Romazini S, Moulin JJ, Pellet F. 2000. Lung cancer mortality in a site
producing hard metals. Occup Environ Med 57(8): 568-573.
457. Witkiewicz-Kucharczyk A, Bal W. 2006. Damage of zinc fingers in DNA repair proteins, a
novel molecular mechanism in carcinogenesis. Toxicol Lett 162(1): 29-42.
458. Wong PK. 1988. Mutagenicity of heavy metals. Bull Environ Contam Toxicol 40(4): 597603. (as cited in IARC 2006)
459. Xia M, Huang R, Sun Y, Semenza GL, Aldred SF, Witt KL, Inglese J, Tice RR, Austin CP.
2009. Identification of chemical compounds that induce HIF-1alpha activity. Toxicol Sci
112(1): 153-163. (Supported by NIEHS and NIH. Authors affiliated with National
Institutes of Health, MD; University of Michigan, MI; Johns Hopkins University School of
Medicine, MD; SwitchGear Genomics, CA; National Institute of Environmental Health
Sciences, NC.)
460. Yamada K. 2013. Cobalt: its role in health and disease. Met Ions Life Sci 13: 295-320.
(Support not reported. Author affiliated with Uniformed Services University of the Health
Sciences, MD.)
461. Yesilada E. 2001. Genotoxicity testing of some metals in the Drosophila wing somatic
mutation and recombination test. Bull Environ Contam Toxicol 66(4): 464-469. (as cited in
IARC 2006)
462. Yildiz M, Cigerci IH, Konuk M, Fidan AF, Terzi H. 2009. Determination of genotoxic
effects of copper sulphate and cobalt chloride in Allium cepa root cells by chromosome
aberration and comet assays. Chemosphere 75(7): 934-938. (Support not reported. Authors
affiliated with Afyon Kocatepe University, Turkey.)
463. Yin DZ. 1990. [Comparison of serum trace element spectrum of liver cancer patients and
healthy adults]. Zhonghua Zhong Liu Za Zhi 12(3): 200-202.
464. Yokoiyama A, Kada T, Kuroda Y. 1990. Antimutagenic action of cobaltous chloride on
radiation-induced mutations in cultured Chinese hamster cells. Mutat Res 245(2): 99-105.
(as cited in IARC 2006)
465. Zeiger E, Anderson B, Haworth S, Lawlor T, Mortelmans K. 1992. Salmonella
mutagenicity tests: V. Results from the testing of 311 chemicals. Environ Mol Mutagen 19
Suppl 21: 2-141. (as cited in IARC 2006)
This draft document should not be construed to represent final NTP determination or policy
181
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
466. Zeller WJ. 1975. [Investigation of the influence of copper and cobalt ions on the
carcinogenesis by diethyl-nitrosamine (DENA) in Wistar rats (author's transl)]. Arch
Geschwulstforsch 45(7): 634-636.
467. Zhang H, Ji Z, Xia T, Meng H, Low-Kam C, Liu R, Pokhrel S, Lin S, Wang X, Liao YP,
Wang M, Li L, Rallo R, Damoiseaux R, Telesca D, Madler L, Cohen Y, Zink JI, Nel AE.
2012b. Use of metal oxide nanoparticle band gap to develop a predictive paradigm for
oxidative stress and acute pulmonary inflammation. ACS Nano 6(5): 4349-4368.
468. Zhang L, Lv J, Sun S. 2012a. Elements in lung tissues of patients from a high lung cancer
incidence area of China. Biol Trace Elem Res 148(1): 7-10.
469. Zhang Q, Kusaka Y, Donaldson K. 2000. Comparative pulmonary responses caused by
exposure to standard cobalt and ultrafine cobalt. J Occup Health 42: 179-184.
470. Zhu S, Zhang Y, Gao Y, Wang XJ, Chen T, Yang Y, Shi R, Jin P, Tian Y, Shen XM. 2011.
[Correlation between level of metallic elements in urine and childhood acute leukemia].
Zhonghua Yu Fang Yi Xue Za Zhi 45(2): 146-149. (Support unknown due to foreign
language. Authors affiliated with Xinhua Hospital affiliated to Shanghai Jiao Tong
University School of Medicine, China.)
471. Zou W, Yan M, Xu W, Huo H, Sun L, Zheng Z, Liu X. 2001. Cobalt chloride induces
PC12 cells apoptosis through reactive oxygen species and accompanied by AP-1 activation.
J Neurosci Res 64(6): 646-653. (Supported by the Chinese Academy of Sciences. Authors
affiliated with Chinese Academy of Sciences, China.)
182
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Abbreviations
ACGIH:
American Conference of Governmental Industrial Hygienists
ADME:
absorption, distribution, metabolism, and excretion
ANOVA:
analysis of variance
atm:
atmosphere
ATSDR:
Agency for Toxic Substances and Disease Registry
bw:
body weight
BDL:
below detection limit
CA:
chromosomal aberration
CASRN:
Chemical Abstracts Service registry number
CDC:
Centers for Disease Control and Prevention
CDR:
Chemical Data Reporting Rule
CI:
confidence interval
CIN:
chromosomal instability
cm2:
centimeters squared
cm3:
centimeters cubed (mL)
DLMI:
dominant lethal mutation index
DLMR:
dominant lethal mutation rate
DNA:
deoxyribonucleic acid
dw:
drinking water
EPA:
Environmental Protection Agency
EQ:
exposure quartiles model
EUSES:
European Union System for the Evaluation of Substances
Exp.:
exposed
F:
female
FDA:
Food and Drug Administration
FR:
Federal Register
ft:
feet
This draft document should not be construed to represent final NTP determination or policy
183
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
FTE:
full-time equivalent
FU:
follow-up
g:
gram
G:
guanine
GC/MS:
gas chromatography/mass spectroscopy
GI:
gastrointestinal
GM:
geometric mean
Hb:
hemoglobin
HETA:
Health Hazard Evaluation and Technical Assistance
HHE:
Health Hazard Evaluation
HHS:
Department of Health and Human Services
HIC:
highest ineffective concentration
HID:
highest ineffective dose
HPLC:
high-performance liquid chromatography
hr:
hour
HWE:
healthy worker effect
HWSE:
healthy worker survival effect
I:
inconclusive
i.m.:
intramuscular
i.p.:
intraperitoneal
i.v.:
intravenous
IARC:
International Agency for Research on Cancer
ICD-7, -8, -9:
International Classification of Diseases, Seventh, Eighth or Ninth Revision
ICD-O
International Classification of Diseases for Oncology
IDLH:
immediately dangerous to life and health
in:
inch
inj.:
injection
JEM:
job-exposure matrix
184
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
kg:
kilogram
L:
liter
LEC:
lowest effective concentration
LED:
lowest effective dose
LOD:
limit of detection
Log Kow:
logarithm of octanol/water partition coefficient
M:
male
m3 :
cubic meter
MCL:
maximum contaminant level
mg:
milligram
mL:
milliliter
MN:
micronuclei
mol:
mole
MS:
mass spectrometry
N:
number
NA
not available; not applicable
NCE:
normochromatic erythrocyte
NCI:
National Cancer Institute
NCTR:
National Center for Toxicological Research
ND:
not detected; not determined; not done
ng:
nanogram
NHANES:
National Health and Nutrition Examination Survey
NI:
no information
NIEHS:
National Institute of Environmental Health Sciences
NIH:
National Institutes of Health
NIOSH:
National Institute for Occupational Safety and Health
NLM:
National Library of Medicine
NOES:
National Occupational Exposure Survey
This draft document should not be construed to represent final NTP determination or policy
6/05/15
185
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
NOS:
not otherwise specified
NPL:
National Priorities List
NR:
not reported; none reported
ns:
not specified
NS:
not significant
NT:
not tested
NTP:
National Toxicology Program
OHAT:
Office of Health Assessment and Translation
OR:
odds ratio
OSHA:
Occupational Safety and Health Administration
P:
probability
P-value:
the statistical probability that a given finding would occur by chance
compared with the known distribution of possible findings
p.o.:
per os (oral administration)
PBZ:
personal breathing zone
PCE:
polychromatic erythrocyte
PEL:
permissible exposure limit
ppm:
parts per million
ppt:
parts per trillion
QSAR:
quantitative structure-activity relationship
R:
estimated daily production of adducts
r:
correlation coefficient
RAHC:
Reasonably anticipated to be a human carcinogen
RBC:
red blood cell
REL:
recommended exposure limit
RNS:
reactive nitrogen species
RoC:
Report on Carcinogens
ROS:
reactive oxygen species
RQ:
reportable quantity
186
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
RR:
relative risk
RTG:
relative total growth
s.c.:
subcutaneous
SAFE:
significance analysis of function and expression
SCE:
sister-chromatid exchange
SD:
standard deviation
SEER:
Surveillance, Epidemiology, and End Results Program, NCI
SIC:
Standard Industrial Classification
SIR:
standardized incidence ratio
SMR:
standardized mortality ratio
SOCMI:
synthetic organic chemical manufacturing industry
SRR:
standardized rate ratio, standardized relative risk
SSB:
single-strand break
STS:
soft tissue sarcoma
TDS:
Total Diet Study
TLV-TWA:
threshold limit value time-weighted average
tmax:
time to maximum concentration in plasma
TMD:
tail moment dispersion coefficient
TRI:
Toxics Release Inventory
TSCA:
Toxic Substances Control Act
TSFE:
time since first employment
UDS:
unscheduled DNA synthesis
UK:
United Kingdom
US:
United States
VOC:
volatile organic compound
WBC:
white blood cell
WHO:
World Health Organization
wk:
week
This draft document should not be construed to represent final NTP determination or policy
6/05/15
187
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
wt%:
weight percent
yr:
year or years
µg:
microgram
188
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Glossary
Ames assay: The Ames Salmonella/microsome mutagenicity assay is a short-term bacterial
reverse mutation assay specifically designed to detect a wide range of chemical substances that
can produce genetic damage that leads to gene mutations.
Analysis bias: A bias arising from inappropriate data assumptions, models, or statistical methods
used to evaluate findings, exposure-response relationships, latency, or confounding.
Aneuploidy: An abnormality involving a chromosome number that is not an exact multiple of
the haploid number (one chromosome set is incomplete).
Apoptosis: Cell deletion by fragmentation into membrane-bound particles, which are
phagocytosed by other cells.
Arabinose resistance: The L-arabinose resistance test with Salmonella typhimurium (Ara test) is
a forward mutation assay that selects a single phenotypic change (from L-arabinose sensitivity to
L-arabinose resistance) in a unique tester strain (an araD mutant).
Aroclor 1254-induced liver: Liver tissue treated with the polychlorinated biphenyl mixture
Aroclor 1254 used as a source of S9 fraction for mutagenic and genotoxic effects testing.
Ascertainment bias: Systematic failure to represent equally all classes of cases or persons
supposed to be represented in a sample.
Attrition bias: Systematic differences between comparison groups in withdrawals or
exclusions of participants from the results of a study.
Biexponential process: A process of drug (or xenobiotic) clearance with two phases with
different rates. The first phase often involves rapid distribution of a drug to peripheral tissues,
while the second phase represents clearance mechanisms that eliminate the drug from the body.
(See “Two-compartment pharmacokinetic model.”)
Boiling point: The boiling point of the anhydrous substance at atmospheric pressure (101.3 kPa)
unless a different pressure is stated. If the substance decomposes below or at the boiling point,
this is noted (dec). The temperature is rounded off to the nearest °C.
Chemical Data Reporting Rule: Chemical Data Reporting (CDR) is the new name for
Inventory Update Reporting (IUR). The purpose of Chemical Data Reporting is to collect quality
screening-level, exposure-related information on chemical substances and to make that
information available for use by the U.S. Environmental Protection Agency (EPA) and, to the
extent possible, to the public. The IUR/CDR data are used to support risk screening, assessment,
priority setting and management activities and constitute the most comprehensive source of basic
screening-level, exposure-related information on chemicals available to EPA. The required
frequency of reporting currently is once every four years.
Co-exposures: substances to which study participants are exposed that can potentially confound
the relationship between the exposure and disease.
Cochran-Armitage trend test: A statistical test used in categorical data analysis when the aim
This draft document should not be construed to represent final NTP determination or policy
189
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
is to assess for the presence of an association between a variable with two categories and a
variable with k categories. It modifies the chi-square test to incorporate a suspected ordering in
the effects of the k categories of the second variable.
Comet assay: The comet assay evaluates DNA damage by measuring DNA migration in single
cells using gel electrophoresis. Migration of DNA is directly related to DNA strand length: the
smaller the strands (produced by breaks in the DNA, i.e., damage), the further the DNA will
migrate from the nucleus in an electric field.
Confounding bias and potential confounders: A bias arising when the comparison groups
under study (e.g., exposed versus unexposed, or the cases versus controls) have different
background risks of disease (Pearce et al. 2007), in effect mixing the association of interest with
the effects of other factors. Potential confounders can include any co-exposures or risk factors
associated with both the exposure and the disease, and that are not part of the disease pathway.
Conversion factor: A numerical factor used to multiply or divide a quantity when converting
from one system of units to another.
Critical temperature: The temperature at and above which a gas cannot be liquefied, no matter
how much pressure is applied.
Differential misclassification bias: A bias that arises when the probability of being
misclassified differs across groups of study subjects. The effect(s) of such misclassification can
vary from an overestimation to an underestimation of the true value.
Differential selection: Selective pressure for self renewal. Gene mutations that confer a growth
or survival advantage on the cells that express them will be selectively enriched in the genome of
tumors.
Disposition: The description of absorption, distribution, metabolism, and excretion of a chemical
in the body.
Dominant lethal mutation assay: The dominant lethal assay identifies germ cell mutagens by
measuring the ability of a chemical to penetrate gonadal tissue and produce embryonic death due
to chromosomal breakage in parent germ cells.
Ecological study: A study in which the units of analysis are populations or groups of people
rather than individuals.
ELISA assay: Enzyme-linked immunosorbent assay; a sensitive immunoassay that uses an
enzyme linked to an antibody or antigen as a marker for the detection of a specific protein,
especially an antigen or antibody.
Epigenetic mechanisms: Changes in gene function that do not involve a change in DNA
sequence but are nevertheless mitotically and/or meiotically heritable. Examples include DNA
methylation, alternative splicing of gene transcripts, and assembly of immunoglobulin genes in
cells of the immune system.
Exposure-response gradient: describes the change in effect caused by differing levels of
exposure (or doses) to a chemical or substance.
190
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
FDA Good Laboratory Practice Regulations: A quality system codified by the U.S. Food and
Drug Administration that prescribes operating procedures for conducting nonclinical laboratory
studies that support or are intended to support applications for research or marketing permits for
products regulated by the Food and Drug Administration.
Fisher’s exact test: The test for association in a two-by-two table that is based on the exact
hypergeometric distribution of the frequencies within the table.
Follow-up: Observation over a period of time of a person, group, or initially defined population
whose appropriate characteristics have been assessed to observe changes in health status or
health-related variables.
Genomic instability: An increased propensity for genomic alterations that often occurs in cancer
cells. During the process of cell division (mitosis) the inaccurate duplication of the genome in
parent cells or the improper distribution of genomic material between daughter cells can result
from genomic instability.
Genotoxic: The property of a chemical or agent that can cause DNA or chromosomal damage.
Healthy worker hire effect: Initial selection of healthy individuals at time of hire so that their
disease risks differ from the disease risks in the source (general) population.
Healthy worker survival effect: A continuing selection process such that those who remain
employed tend to be healthier than those who leave employment.
Henry’s Law constant: The ratio of the aqueous-phase concentration of a chemical to its
equilibrium partial pressure in the gas phase. The larger the Henry’s law constant the less soluble
it is (i.e., greater tendency for vapor phase). The relationship is defined for a constant
temperature, e.g., 25°C.
Information bias: a bias arising from measurement error. Information bias is also referred to as
observational bias and misclassification (see differential and non-differential misclassification
bias). When any exposure, covariate, or outcome variable is subject to measurement error, a
different quality or accuracy of information between comparison groups can occur.
Integration of scientific evidence across studies: the final step in the cancer assessment that
assigns greater weight to the most informative studies to reach a preliminary listing
recommendation.
Job exposure matrix (JEM): a tool used to assess exposure to potential health hazards in
occupational epidemiologic studies by converting coded occupational data (usually job titles)
into a matrix of possible levels of exposures to potentially harmful agents, reducing the need to
assess each individual's exposure in detail.
Lagging: Statistical methods that weight exposure times in order to account for prolonged
induction and latency periods, particularly in occupational epidemiology studies.
Latency and prolonged induction: The induction period is the time required for a cause to lead
to the disease process (regardless of symptoms); the latent period is the time between the
This draft document should not be construed to represent final NTP determination or policy
191
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
exposure and clinical manifestation of the disease. Especially important when considering cancer
outcomes.
Left truncation: This bias can occur when workers hired before the start of the study, and thus
exposed and at risk for disease, do not remain observable at the start of follow-up. The
remaining prevalent workers may be healthier and not representative of all workers hired before
the start of the study.
Melting point: The melting point of the substance at atmospheric pressure (101.3 kPa). When
there is a significant difference between the melting point and the freezing point, a range is
given. In case of hydrated substances (i.e., those with crystal water), the apparent melting point is
given. If the substance decomposes at or below its melting point, this is noted (dec). The
temperature is rounded off to the nearest °C.
Metaplasia: A change of cells to a form that does not normally occur in the tissue in which it is
found.
Methemoglobin: A form of hemoglobin found in the blood in small amounts. Unlike normal
hemoglobin, methemoglobin cannot carry oxygen. Injury or certain drugs, chemicals, or foods
may cause a higher-than-normal amount of methemoglobin to be made. This causes a condition
called methemoglobinemia.
Micronuclei: Small nuclear-like bodies separate from, and additional to, the main nucleus of a
cell, produced during the telophase of mitosis or meiosis by lagging chromosomes or
chromosome fragments derived from spontaneous or experimentally induced chromosomal
structural changes.
Miscible: A physical characteristic of a liquid that forms one liquid phase with another liquid
(e.g., water) when they are mixed in any proportion.
Molecular weight: The molecular weight of a substance is the weight in atomic mass units of all
the atoms in a given formula. The value is rounded to the nearest tenth.
Mutagenic: Capable of inducing genetic mutation, e.g., a genotoxic substance or agent that can
induce or increase the frequency of mutation in the DNA of an organism.
Mutations: A change in the structure of a gene, resulting from the alteration of single base units
in DNA, or the deletion, insertion, or rearrangement of larger sections of genes or chromosomes.
The genetic variant can be transmitted to subsequent generations.
National Health and Nutrition Examination Survey: A program of studies designed to assess
the health and nutritional status of adults and children in the United States. The survey is unique
in that it combines interviews and physical examinations.
Nondifferential misclassification bias: arises when all classes, groups, or categories of a
variable (whether exposure, outcome, or covariate) have the same error rate or probability of
being misclassified for all study subjects. In the case of binary or dichotomous variables
nondifferential misclassification would usually result in an 'underestimation' of the hypothesized
relationship between exposure and outcome.
192
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
Normochromatic erythrocyte: A mature erythrocyte that lacks ribosomes and can be
distinguished from immature, polychromatic erythrocytes by stains selective for RNA.
Octanol/water partition coefficient (log Kow): A measure of the equilibrium concentration of a
compound between octanol and water.
One-compartment model: A pharmacokinetic modeling approach that models the entire body
as a single compartment into which a drug is added by a rapid single dose, or bolus. It is assumed
that the drug concentration is uniform in the body compartment at all times and is eliminated by
a first order process that is described by a first order rate constant.
Personal breathing zone: A sampling area as close as practical to an employee’s nose and
mouth, (i.e., in a hemisphere forward of the shoulders within a radius of approximately nine
inches) so that it does not interfere with work performance or safety of the employee.
Personal protective equipment: Specialized clothing or equipment, worn by an employee to
minimize exposure to a variety of hazards. Examples of PPE include such items as gloves, foot
and eye protection, protective hearing devices (earplugs, muffs) hard hats, respirators and full
body suits.
Plate incorporation: A commonly used procedure for performing a bacterial reverse mutation
test. Suspensions of bacterial cells are exposed to the test substance in the presence and in the
absence of an exogenous metabolic activation system. In the plate-incorporation method, these
suspensions are mixed with an overlay agar and plated immediately onto minimal medium. After
two or three days of incubation, revertant colonies are counted and compared with the number of
spontaneous revertant colonies on solvent control plates.
Point emission: A release that can be identified with a single discharge source or attributed to a
specific physical location.
Poly-3 trend test: A survival-adjusted statistical test that takes survival differences into account
by modifying the denominator in the numerical (quantal) estimate of lesion incidence to reflect
more closely the total number of animal years at risk.
Proto-oncogene: A gene involved in normal cell growth. Mutations (changes) in a protooncogene may cause it to become an oncogene, which can cause the growth of cancer cells.
Proxy: a substitute authorized to act for the study participant. Often this is a spouse or other
family member who may consent to be interviewed, offering information about the participant.
Ptrend: Level of statistical significance of a change over time in a group selected to represent a
larger population.
QUOSA: A collection of scientific literature management software and services for researchers
and information professionals in the life sciences and related scientific and medical areas
designed to retrieve, organize, and analyze full-text articles and documents.
Recall bias: a bias arising from systematic error in the accuracy or completeness of "recalled" by
study participants regarding past events, and usually arises in the context of retrospective casecontrol interviews or questionnaires. The concern is that those with the disease may search their
This draft document should not be construed to represent final NTP determination or policy
193
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
memories more thoroughly than unaffected controls to try to recall exposure to various causal
factors. This bias is often differential and biases towards an overestimate of effect.
Reverse causality: may arise in case-control studies when exposure is measured after disease
diagnosis, as the concern is that symptoms or early manifestations of the disease may affect the
measured exposure; this is particularly of concern in studies using biomarkers of effect.
Right truncation: for right truncated data, only participants or person-time under observation up
to a given date are included. Right truncation results in limiting person-time to values that are
limited below the given date. Truncation is similar to but distinct from the concept statistical
censoring. A truncated sample is similar to an underlying sample with all values outside the
bounds entirely omitted, with no count of participants or person-time omitted kept. Alternatively,
with statistical censoring, the value of the bound exceeded is known and documented.
Selection bias: An error in choosing the individuals or groups to take part in a study. Ideally, the
subjects in a study should be very similar to one another and to the larger population from which
they are drawn (for example, all individuals with the same disease or condition). If there are
important differences, the results of the study may not be valid, and bias can be introduced in
either direction.
Selective reporting: selective reporting occurs when the effect estimate for a measurement (of
exposure or disease) was selected from among analyses using several measurement instruments,
reflecting the most favorable result or subcategories.
Sensitivity: the proportion of truly diseased persons in the screened population who are
identified as diseased by the screening test; or the probability of correctly diagnosing a true case
with the test.
Sister-chromatid exchange: The exchange during mitosis of homologous genetic material
between sister chromatids; increased as a result of inordinate chromosomal fragility due to
genetic or environmental factors.
Solubility: The ability of a substance to dissolve in another substance and form a solution. The
Report on Carcinogens uses the following definitions (and concentration ranges) for degrees of
solubility: (1) miscible (see definition), (2) freely soluble- capable of being dissolved in a
specified solvent to a high degree (> 1,000 g/L), (3) soluble- capable of being dissolved in a
specified solvent (10–1,000 g/L), (4) slightly soluble- capable of being dissolved in a specified
solvent to a limited degree (1-10 g/L), and (5) practically insoluble- incapable of dissolving to
any significant extent in a specified solvent (< 1 g/L).
Specific gravity: The ratio of the density of a material to the density of a standard material, such
as water at a specific temperature; when two temperatures are specified, the first is the
temperature of the material and the second is the temperature of water.
Specificity: the proportion of truly nondiseased persons who are so identified by the screening
test; or the probability of correctly identifying a non-diseased person with the test.
Spot test: Qualitative assay in which a small amount of test chemical is added directly to a
selective agar medium plate seeded with the test organism, e.g., Salmonella. As the chemical
diffuses into the agar, a concentration gradient is formed. A mutagenic chemical will give rise to
194
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
6/05/15
a ring of revertant colonies surrounding the area where the chemical was applied; if the chemical
is toxic, a zone of growth inhibition will also be observed.
Study sensitivity: the ability of a study to detect an effect (if it exists) which would include a
large number of exposed cases; evidence of substantial exposure (e.g., level, duration, frequency,
or probability) during an appropriate window; an adequate range in exposure levels or duration
allowing for evaluation of exposure-response relationships; and an adequate length of follow-up.
Study utility: the overall utility of a study is based on consideration of the potential for bias (i.e.,
study quality) and study sensitivity. Serious concerns about study quality will result in lower
utility of the study; a high quality study with low sensitivity could also have low utility.
Surrogate exposure data: ideally, a study would provide multiple quantitative metrics of each
individual’s exposure to the substance of interest. However, a surrogate metric correlated with
exposure may be used instead of, or in addition to exposure data.
Time-weighted average: The average exposure concentration of a chemical measured over a
period of time (not an instantaneous concentration).
Toxicokinetics: The mathematical description (toxicokinetic models) of the time course of
disposition of a chemical in the body.
Transitions: DNA nucleotide substitution mutation in which a purine base is substituted for
another purine base (adenine → guanine or guanine → adenine) or a pyrimidine base for another
pyrimidine base (cytosine → thymine or thymine → cytosine).
Transversions: DNA nucleotide substitution mutation in which a purine base (adenine or
guanine) is substituted for a pyrimidine base (cytosine or thymine) or vice versa.
Two-compartment pharmacokinetic model: A two-compartment pharmacokinetic model
resolves the body into a central compartment and a peripheral compartment. The central
compartment generally comprises tissues that are highly perfused such as heart, lungs, kidneys,
liver and brain. The peripheral compartment comprises less well-perfused tissues such as muscle,
fat and skin. A two-compartment model assumes that, following drug administration into the
central compartment, the drug distributes between that compartment and the peripheral
compartment. However, the drug does not achieve instantaneous distribution (i.e., equilibrium),
between the two compartments. After a time interval (t), distribution equilibrium is achieved
between the central and peripheral compartments, and elimination of the drug is assumed to
occur from the central compartment.
Type-I error: The error of rejecting a true null hypothesis, i.e., declaring that a difference exists
when it does not.
Type-II error: The error of failing to reject a false null hypothesis, i.e., declaring that a
difference does not exist when in fact it does.
Vapor density, relative: A value that indicates how many times a gas (or vapor) is heavier than
air at the same temperature. If the substance is a liquid or solid, the value applies only to the
vapor formed from the boiling liquid.
This draft document should not be construed to represent final NTP determination or policy
195
6/05/15
Peer-Review Draft: Report on Carcinogens Monograph on Cobalt
Vapor pressure: The pressure of the vapor over a liquid (and some solids) at equilibrium,
usually expressed as mm Hg at a specific temperature (°C).
196
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
Part 2
Draft Profile
This draft document should not be construed to represent final NTP determination or policy
II-1
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
This Page Intentionally Left Blank
II-2
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
Cobalt and Certain Cobalt Compounds
CAS No. 7440-48-4 (Cobalt metal)
No separate CAS No. assigned for cobalt compounds as a class
Reasonably anticipated to be human carcinogens
Introduction
The compound cobalt sulfate was first listed in the Eleventh Report on Carcinogens in 2004 as
reasonably anticipated to be a human carcinogen based on sufficient evidence of carcinogenicity
in experimental animals. The listing of cobalt and certain cobalt compounds supersedes the
previous listing of cobalt sulfate in the Report on Carcinogens and applies to cobalt and certain
cobalt compounds as defined below.
Carcinogenicity
Cobalt and certain cobalt compounds are reasonably anticipated to be human
carcinogens based on sufficient evidence of carcinogenicity from studies in experimental
animals and supporting data from studies on mechanisms of carcinogenesis. “Certain cobalt
compounds” are defined as compounds that release cobalt ions in vivo, which mechanistic data
indicate is a key event for cobalt-induced carcinogenicity. The available data show that cobalt
metal and certain cobalt compounds (regardless of their solubility in water) act via similar modes
of action and induce similar cytotoxic, genotoxic, and carcinogenic effects, and that the cobalt
ion is largely responsible for the toxicity and carcinogenicity (NTP 1998, 2014, IARC 2006).
Both water-soluble cobalt compounds, which release ions in extracellular fluids, and poorly
water-soluble cobalt particles, which release cobalt ions intracellularly in lysosomes, are
included in this grouping. Cobalt metal and all of the cobalt water-soluble and poorly watersoluble cobalt compounds that have been evaluated have been found to be soluble in biological
fluids (e.g., gastric and lysosomal fluids), as discussed under “Properties” below, suggesting that
they will release cobalt ions in vivo (Hillwalker and Anderson 2014, Brock and Stopford 2003,
Stopford et al. 2003). Vitamin B12, which is an essential cobalt-containing nutrient, does not
meet the criteria for “certain cobalt compounds,” because it does not release cobalt ions and
passes through the body intact while bound to specific carrier proteins (Neale 1990).
Mechanisms of Carcinogenesis and Other Relevant Data
The key events related to toxicity and carcinogenicity are thought to include cellular uptake of
cobalt, intracellular release of cobalt ions from particles, and immediate and downstream
biological responses related to the proposed modes of action (as shown in the diagram below).
The first step in the carcinogenicity or toxicity process is the release of cobalt ions in vivo.
Water-soluble cobalt compounds release ions into extracellular fluids, and poorly water-soluble
cobalt particles release cobalt ions intracellularly in lysosomes.
This draft document should not be construed to represent final NTP determination or policy
II-3
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
Mechanistic events in cobalt carcinogenicity
Several key events have been identified that are related to biologically plausibile modes of
actions and are applicable to all cobalt forms that release cobalt ions in vivo. These events
include inhibition of DNA repair, genotoxicity, generation of reactive oxygen species (ROS) and
oxidative damage, and stabilization of hypoxia-inducible factor 1α (HIF-1α).
Cobalt is mutagenic in mammalian cells and induces DNA strand breaks and chromosome
damage in vitro. Only a few in vivo genotoxicity studies were available, but the results were
generally consistent with those of in vitro studies. Although the mechanisms of cobalt-induced
genetic damage are not completely understood, the literature suggests two possible mechanisms:
(1) a direct effect of cobalt(II) ions to induce oxidative damage to DNA, and/or (2) an indirect
effect through inhibition of DNA repair (Smith et al. 2014, Lison 2015).
Cobalt is also a redox-active transition metal, and in vitro studies have shown that cobalt
particles and ions can induce ROS in mammalian cells, with cobalt metal and cobalt oxide
particles having a greater effect than ions. Evidence of oxidative stress and oxidative damage
have been shown in in vivo studies in rat kidney, liver, and lung. Also, a higher frequency of G to
T transversion mutations in the K-ras oncogene (a common mutation associated with oxidative
DNA damage) were found in cobalt induced lung tumors in mice and rats compared to
spontaneous lung tumors (NTP 1998, 2014, IARC 2006). In addition to directly inducing DNA
damage, ROS also activate a number of redox-sensitive transcription factors (e.g., nuclear factor
κB, activator protein 1, and tumor protein p53) that have been linked to carcinogenesis because
of their role in regulating inflammation, cell proliferation, differentiation, angiogenesis, and
apoptosis (Valko et al. 2005, 2006, Beyersmann and Hartwig 2008). Thus, ROS may initiate
tumor development by mutagenesis and/or promote tumor growth by dysregulation of cell
growth and proliferation.
Finally, a well-established biological effect of cobalt is to mimic hypoxia by stabilizing
HIF-1α (Maxwell and Salnikow 2004, Greim et al. 2009, Saini et al. 2010a,b, Galán-Cobo et al.
2013, Gao et al. 2013, Nyga et al. 2015). HIF-1α plays a central role in the transcriptional
regulation of more than 100 hypoxia-responsive genes and is a major regulator of the adaptation
of cancer cells to hypoxia. HIF-1α overexpression has been linked to cancer initiation and
II-4
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
progression and is a common characteristic of many human cancers (Paul et al. 2004, Galanis et
al. 2008, 2009, Cheng et al. 2013).
Although most of the toxicological effects of cobalt are attributed to the cobalt ion, direct
toxic effects of cobalt particles also contribute, as evidenced by the greater toxicity of cobalt
metal than of cobalt sulfate in National Toxicology Program (NTP) rodent bioassays (NTP 1998
2014, Behl et al. 2015). Differences in the relative toxicity reported for cobalt particles and ions
may be partially explained by differences in cellular uptake mechanisms, a synergistic effect
between the particles and metal on ROS production, and differences in intracellular cobalt
accumulation and distribution (Peters et al. 2007, Sabbioni et al. 2014, Smith et al. 2014).
Cancer Studies in Experimental Animals
Exposure of experimental animals to cobalt metal or cobalt compounds caused tumors in two
rodent species, at several different tissue sites, and by several different routes of exposure. This
conclusion is based on studies in rats and mice exposed to cobalt metal (five studies), watersoluble cobalt compounds (two studies with cobalt sulfate and one study with cobalt chloride),
and poorly water-soluble cobalt compounds (four studies with cobalt oxide). Studies of cobalt
alloys and radioactive cobalt in experimental animals were not considered to be informative
because of potential confounding by other carcinogens.
Inhalation exposure of rats and mice to cobalt metal (NTP 2014) or cobalt sulfate (NTP
1998) or intratracheal instillation of cobalt oxide in rats (Steinhoff and Mohr 1991) caused lung
tumors (alveolar/bronchiolar adenoma and carcinoma). In addition, inhalation exposure of rats to
cobalt metal caused squamous-cell tumors of the lung (primarily cystic keratinizing epithelioma)
in females and possibly in males.
In inhalation studies of cobalt metal in rats, tumors were also induced at sites distant from
the lung, including tumors of the pancreas (islet-cell adenoma or carcinoma combined) in males
and of the hematopoietic system (mononuclear-cell leukemia) in females, indicating a systemic
effect (NTP 2014). Increased incidence of neoplasms in the kidney (adenoma or carcinoma
combined) in male rats and pancreas (carcinoma) in female rats may have been related to cobalt
metal inhalation (NTP 2014). Exposure to cobalt metal or cobalt sulfate induced adrenal gland
tumors (benign and malignant pheochromocytoma). The inhalation-exposure studies conducted
with cobalt metal and cobalt sulfate suggested that cobalt metal was more carcinogenic than
cobalt sulfate at a similar cobalt concentration, based on the incidence and spectrum of lung
tumors and the extent of systemic lesions (Behl et al. 2015). This finding is consistent with the
greater cytotoxicity of and ROS induction for cobalt metal in the in vitro mechanistic studies
described above.
In rats, local injection of cobalt at various anatomic locations caused tumors at the injection
sites. Although these studies were less robust than the inhalation studies and sarcomas are
common in injection studies in rats on a variety of compounds, the consistency of the tumor
types and findings across different cobalt forms provide supporting evidence of carcinogenicity
of cobalt. Intraperitoneal or intramuscular injection of the poorly water-soluble compound cobalt
oxide caused histiocytoma and/or sarcoma at the injection site (Gilman and Ruckerbauer 1962,
Steinhoff and Mohr 1991), and subcutaneous injection of the water-soluble compound cobalt
chloride caused fibrosarcoma (Shabaan et al. 1977). Intramuscular or intrathoracic injection of
cobalt metal (Heath 1956, Heath and Daniel 1962) or nanoparticles (Hansen et al. 2006) caused
sarcoma (primarily rhabdomyofibrosarcoma, rhabdomyosarcoma, or fibrosarcoma
This draft document should not be construed to represent final NTP determination or policy
II-5
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
A few studies in rodents (Gilman and Ruckerbauer 1962, Jasmin and Riopelle 1976,
Wehner et al. 1977) found no tumors at certain tissue sites following exposure to the same forms
of cobalt that caused tumors in other studies; however, these studies generally lacked sensitivity
to detect an effect, because of the use of a less sensitive animal model, shorter study duration, or
lower exposure levels.
Cancer Studies in Humans
The data available from studies in humans are inadequate to evaluate the relationship between
human cancer and exposure specifically to cobalt and certain cobalt compounds. The data
relevant to the evaluation were from studies of five independent cohorts of workers, primarily
evaluating lung cancer, and two population-based case-control studies of esophageal and other
cancers of the aerodigestive tract, one in Ireland (O'Rorke et al. 2012) and the other in the state
of Washington (Rogers et al. 1993). The cohorts included (1) porcelain painters in Denmark
(Tüchsen et al. 1996), (2) cobalt production workers in an electrochemical plant in France
reported in two publications (Mur et al. 1987, Moulin et al. 1993), (3) two overlapping cohorts
of cobalt–tungsten carbide hard-metals workers in France (Moulin et al. 1998, Wild et al. 2000),
(4) stainless- and alloyed-steel workers in France (Moulin et al. 2000), and (5) nickel refinery
workers in Norway (Grimsrud et al. 2005). Studies of cobalt alloys in humans (primarily metalon-metal implants) were not considered to be informative, because they were not specific to
cobalt exposure.
Although increased risks of lung cancer were found in most of the cohort studies, and
increases in esophageal cancer were suggested in the two case-control studies, it is unclear that
the excess risks were due to exposure specifically to cobalt, because of potential confounding
from exposure to known lung carcinogens or other study limitations. In the cohort studies, hardmetal (Moulin et al. 1998, Wild et al. 2000) and nickel refinery workers (Grimsrud et al. 2005)
were also exposed to known lung carcinogens; excess risks were also found among the
“unexposed” referent pottery workers; and the excess risk found in an earlier cohort study of
cobalt production workers (Mur et al. 1987) was no longer present in a later update of the cohort
(Moulin et al. 1993). In the case-control studies, cobalt exposure was assessed in toenail samples
taken at or after diagnosis of esophageal cancer. It therefore is unclear whether cobalt levels in
the toenails reflected a relevant period of exposure, or whether cobalt exposure preceded cancer
or resulted from changes due to tumor formation.
Properties
Cobalt and certain cobalt compounds as a class are related largely by their chemical properties,
specifically bioavailability.
Bioavailability
Because the carcinogenic and toxic effects of cobalt and certain cobalt compounds begin with the
release of cobalt ions in vivo, the bioavailability of cobalt ions is critical for consideration of
carcinogenicity. The bioavailability of a metal species is determined by its solubility in
biological fluids. Bioaccessibility studies (testing solubility in synthetic biological fluids) have
demonstrated that cobalt metal and both water-soluble and poorly water-soluble cobalt
compounds can dissolve and release cobalt ions in some biological fluids (Brock and Stopford
2003, Stopford et al. 2003), suggesting that they will release ions in vivo. In vitro bioaccessibility
studies using synthetic equivalents of gastric fluid (for ingestion exposure) and lysosomal fluids
II-6
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
(for inhalation exposure) (Brock and Stopford 2003, Stopford et al. 2003, Hillwalker and
Anderson 2014) confirmed the complete (or mostly complete) solubility of cobalt metal,
representative water-soluble cobalt compounds (cobalt sulfate heptahydrate and chloride), and
both organic and inorganic poorly water-soluble cobalt compounds (cobalt oxide, 2-ethylhexanoate, carbonate, and naphthenate) in both of these low-pH biological fluids. The
bioaccessibility results are shown in the table below, along with other chemical and physical
properties of cobalt metal and these cobalt compounds. Results with other biological fluids, such
as serum and intestinal, alveolar, and interstitial fluids, indicate that species of cobalt compound,
particle size and surface area, and the pH of the surrogate fluid can affect the solubility of cobalt
in biological fluids.
Physical and chemical properties for cobalt metal and cobalt compounds that have been tested
for bioaccessibility
a
CAS No.
Cobalt metal
7440-48-4
Form
b
Molec.
weight
Formula
Co
e
Bioaccessibility
Density
(% solubility in
or
Water
gastric/
specific solubility
c
d
gravity (g/100 cc) lysosomal fluids)
Physical form
58.9
e
grey hexagonal or
e
cubic metal
281.1
g
red pink, monoclinic
129.8
h
blue hexagonal
h
leaflets
74.9
g
green-brown cubic
173.7
h
blue liquid (12% Co)
118.9
g
red, trigonal
401.3
e
8.92
e
0.000875
f
1.95
g
60.4
3.36
h
45
6.45
g
i
g
100/92.4
1.01
g
28.8
k
(20°C)
100/100
4.13
g
i
0.97
g
100/100
Water-soluble compounds
g
Sulfate
heptahydrate
10026-24-1
CoSO4•7H2O
Chloride
7646-79-9
CoCl2
h
g
g
100/100
h
100/100
Poorly water-soluble compounds
Oxide
1307-96-6
g
CoO
2-Ethylhexanoate
(org.)
136-52-7
Co(C8H15O2)2
Carbonate (org.)
513-79-1
CoCO3
Naphthenate (org.) 61789-51-3
g
g
Co(C11H7O2)2
e
g
g
g
purple liquid (6% Co)
g
g
34.3 (20°C)
100/100
j
100/100
a
All compounds contain Co(II). Forms in italics have been tested for carcinogenicity, genetic toxicity, or have
mechanistic data; org. = organic compound; all others are inorganic.
b
SciFinder (2015)
c
Solubility in cold water unless otherwise indicated; i = insoluble;
d
e
f
g
h
i
j
Stopford et al. 2003, PubChem 2015, ChemIDplus 2015, CDI 2006, HSBD 2015, ATSDR 2004, MorningStar
k
2005a, MorningStar 2005b.
The solubility of cobalt compounds in water is largely pH dependent, and cobalt is
generally more mobile in acidic solutions than in alkaline solutions (IARC 1991, Paustenbach et
al. 2013). Sulfates, nitrates, and chlorides of cobalt tend to be soluble in water, whereas oxides
(including the mixed oxide, Co3O4), hydroxides, and sulfides tend to be poorly soluble or
insoluble in water (Lison 2015). Organic cobalt compounds can be either soluble, as with
cobalt(II) acetate, or insoluble, as with cobalt(II) carbonate and cobalt(II) oxalate (CDI 2006). In
addition to low pH, solubilization of some poorly water-soluble compounds in biological fluids
may be enhanced in the presence of binding proteins (IARC 2006).
This draft document should not be construed to represent final NTP determination or policy
II-7
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
Chemical characteristics
Cobalt (Co) is a naturally occurring transition element with magnetic properties. It is the 33rd
most abundant element, making up approximately 0.0025% of the weight of Earth’s crust. Cobalt
is a component of more than 70 naturally occurring minerals, including arsenides, sulfides, and
oxides. The only stable and naturally occurring cobalt isotope is 59Co (ATSDR 2004, WHO
2006). Metallic cobalt, Co(0), exists in two allotropic forms, hexagonal and cubic, which are
stable at room temperature (IARC 1991, ATSDR 2004, WHO 2006). Cobalt predominantly
occurs in two oxidation states, Co(II) and Co(III). Co(II) is much more stable than Co(III) in
aqueous solution (Nilsson et al. 1985, Paustenbach et al. 2013) and is present in the environment
and in most commercially available cobalt compounds (e.g., cobalt chloride, sulfide, and sulfate).
Co(III) is also present in some commercially available cobalt compounds, including the mixed
oxide (Co3O4) (IARC 1991, Paustenbach et al. 2013, Lison 2015) and some simple salts of
Co(III) (e.g., Co2O3). Important salts of carboxylic acids include formate, acetate, citrate,
naphthenate, linoleate, oleate, oxalate, resinate, stearate, succinate, sulfamate, and 2ethylhexanoate.
Use
Cobalt and cobalt compounds are used in numerous commercial, industrial, and military
applications. On a global basis, the largest use of cobalt is in rechargeable battery electrodes. In
2012, the reported U.S. consumption of cobalt and cobalt compounds was approximately
8,420 metric tons, the majority used for superalloys (Shedd 2014b). Major uses for metallic
cobalt include production of superalloys, cemented carbides, and bonded diamonds. Cobalt
nanoparticles are used in medical applications (e.g., sensors, magnetic resonance imaging
contrast enhancement, drug delivery), and cobalt nanofibers and nanowires are used in industrial
applications. Cobalt compounds are used as pigments for glass, ceramics, and enamels (oxides,
sulfate, and nitrate), as driers for paints, varnishes, or lacquers (hydroxide, oxides, propionate,
acetate, tallate, naphthenate, and 2-ethylhexanoate), as catalysts (hydroxide, oxides, carbonate,
nitrate, acetate, oxalate, and sulfide), as adhesives and enamel frits (naphthenate, stearate, and
oxides), and as trace mineral additives in animal diets (carbonate, sulfate, nitrate, oxides, and
acetate). U.S. consumption of cobalt and cobalt compounds in 2012 is summarized in the
following table.
Metric tons of
cobalt content
End use
Percent of total
consumption
Superalloys
4,040
48.0
Chemicals and ceramics
2,300
27.3
774
9.2
699
8.3
548
6.5
63
0.7
Cemented carbides
Other alloys
a
Steels
Miscellaneous and
unspecified
Source: Shedd 2014b.
a
Includes magnetic, nonferrous, and wear-resistant alloys and welding materials.
II-8
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
The fastest-growing use for cobalt in recent years has been in high-capacity, rechargeable
batteries, including nickel-cadmium, nickel-metal hydride, and lithium-ion batteries for electric
vehicles and portable electronic devices such as smart phones and laptops (Maverick 2015).
Many other uses for cobalt exist, including in integrated circuit contacts and semiconductor
production. An emerging area of use is as a key element in several forms of “green” energy
technology applications, including gas-to-liquids and coal-to-liquids processes, oil
desulfurization, clean coal, solar panels, wind and gas turbines, and fuel cells, and in cobaltbased catalysts for sunlight-driven water-splitting to convert solar energy into electrical and
chemical energy.
Production
Cobalt metal is produced as a by-product from ores associated with copper, nickel, zinc, lead,
and platinum-group metals and is most often chemically combined in its ores with sulfur and
arsenic (Davis 2000, CDI 2006). The largest cobalt reserves are in the Congo (Kinshasa),
Australia, Cuba, Zambia, Canada, Russia, and New Caledonia, with very limited production in
the United States in recent years (Shedd 2014a). Except for a negligible amount of by-product
cobalt produced from mining and refining of platinum-group metal ores, the United States did
not refine cobalt in 2012 (Shedd 2014b). Cobalt has not been mined in the United States in over
30 years (ATSDR 2004); however, a primary cobalt mine, mill, and refinery were being
established in Idaho in 2015 (Farquharson 2015). In 2012, 2,160 metric tons of cobalt was
recycled from scrap. No cobalt has been sold from the National Defense Stockpile since 2009.
Metallic cobalt and several cobalt compounds are high-production-volume (HPV)
chemicals, based on their annual production or importation into the United States in quantities of
at least 1 million pounds. Recent volumes of U.S. production, imports, and exports of cobalt
metal and HPV cobalt compounds are listed in the following table.
Quantity (lb)
Cobalt category
Production
(2012)
Imports
(2013)
Exports
(2013)
Metal (excluding alloys)
23,384,002
16,151,599
–
1 million to < 10 million
342,918
520,996
1,038,821
1,193,856
–
215,661
14,304
–
a
Compounds:
Acetates
Carbonates
Chlorides
–
b
2-Ethylhexanoate
4,294,523
–
Hydroxide
4,709,137
–
a
–
c
Oxides
1 million to < 10 million
5,300,984
Propionate
1 million to < 10 million
–
–
Sulfate
1 million to < 10 million
1,319,004
–
902,467
c
a
– = no data found.
a
No specific U.S. Census Bureau Schedule B export code was identified.
b
Cobalt chloride production data for 2012 were withheld by the manufacturer.
c
The reported value is for cobalt hydroxide and oxides combined.
This draft document should not be construed to represent final NTP determination or policy
II-9
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
Exposure
A significant number of people living in the United States are exposed to cobalt, based on
several lines of evidence, including biological monitoring data demonstrating exposure in
occupationally and non-occupationally exposed populations. Data from the U.S. Environmental
Protection Agency’s Toxics Release Inventory (TRI) indicate that production- and use-related
releases of cobalt compounds have occurred at numerous industrial facilities in the United States.
In biomonitoring studies that measured cobalt in the urine of people exposed to cobalt from
various sources, the highest levels generally were due to occupational exposures and failed hip
implants; lower levels were due to exposure from normal implants or the environment. Low
levels were also observed in the general population (with unknown sources of exposure). The
following graph shows the mean or median levels of urinary cobalt for the general public and for
groups with known exposures. Data are reported for both U.S. and non-U.S. exposures;
occupational and medical implant exposures outside the United States can be informative
because of the similar production methods and implant compositions worldwide.
Urine levels of cobalt for various exposed groups
Filled symbols = U.S. data; open symbols = non-U.S. data.
Urinary cobalt measurements in the U.S. general public have remained consistent since
1999, with geometric mean values between 0.316 and 0.379 µg/L, according to the National
Health and Nutrition Examination Survey (NHANES) (CDC 2015). Urinary cobalt is considered
a good indicator of absorbed cobalt (IARC 2006, WHO 2006), especially from recent exposures
(ATSDR 2004). Levels of cobalt in blood (including whole blood, plasma, and serum) show a
pattern similar to that for urinary cobalt levels.
Occupational exposure
The primary route of occupational exposure to cobalt is via inhalation of dust, fumes, or mists or
gaseous cobalt carbonyl. Dermal contact with cemented carbide (i.e., hard-metal) powders and
cobalt salts can result in systemic uptake. Occupational exposure to cobalt occurs during refining
II-10
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
of cobalt; production and use of cobalt alloys; hard-metal production, processing, and use;
maintenance and re-sharpening of hard-metal tools and blades; manufacture and use of cobaltcontaining diamond tools; and use of cobalt-containing pigments and driers. Workers
regenerating spent catalysts may also be exposed to cobalt sulfides. Occupational exposure has
been documented by measurements of cobalt in ambient workplace air and in blood, urine, nails,
and hair, and lung tissue from workers or deceased workers (IARC 1991, ATSDR 2004, IARC
2006, CDC 2013). The highest levels of cobalt in workplace air are generally for hard-metal
manufacture involving cobalt metal powders (1,000 to 10,000 µg/m3) (NTP 2009) and for
production of cobalt acetate, chloride, nitrate, oxide, and sulfate (IARC 2006).
The National Institute of Occupational Safety and Health (NIOSH) National Occupational
Exposure Survey (conducted from 1981 to 1983) estimated that approximately 386,500 workers
were potentially exposed to cobalt and cobalt compounds (NIOSH 1990).
Surgical implants
As mentioned above, cobalt implants are a major source of exposure to cobalt in patients
receiving orthopedic joint replacements, especially hip implants, which contain cobalt in cobaltchromium-molybdenum alloys (Sampson and Hart 2012, Devlin et al. 2013). Implants may fail
because of excessive wear or corrosion by body fluids, increasing the levels of cobalt released
from the implants (Sampson and Hart 2012). A recommended level of blood cobalt for further
clinical investigation and action has been set at 7 µg/L in the United Kingdom (MHRA 2012)
and at 10 µg/L in the United States by the Mayo Clinic (2015).
Environmental exposure
Evidence of the potential for environmental exposure to cobalt comes from biomonitoring
studies that found elevated levels of cobalt in people who lived near mining operations in
Guatemala (Basu et al. 2010) and Mexico (Moreno et al. 2010). The TRI reported that in 2013,
on- and off-site industrial releases of cobalt and cobalt compounds totalled approximately
5.5 million pounds from 723 facilities in the United States (TRI 2014a). Calculations based on
media-specific release data from TRI indicate that releases to land accounted for 82% of total
releases in 2013. Worldwide, approximately 75,000 metric tons of cobalt enters environment
annually (Shedd 1993, CDI 2006) with similar amounts coming from natural sources (40,000
metric tons) and anthropogenic sources (35,000 metric tons) (Shedd 1993, CDI 2006, TRI
2014b).
The average concentration of cobalt in ambient air in the United States has been reported to
be approximately 0.4 ng/m3 (ATSDR 2004). Levels can be orders of magnitude higher near
source areas (e.g., near facilities processing cobalt-containing alloys and compounds) reported
from outside the United States. The median cobalt concentration in U.S. drinking water has been
reported to be less than 2.0 µg/L; however, levels as high as 107 µg/L have been reported
(ATSDR 2004). Cobalt concentrations have been reported to range from 0.01 to 4 µg/L in
seawater and from 0.1 to 10 µg/L in fresh water and groundwater (IARC 2006). Studies have
reported cobalt soil concentrations ranging from 0.1 to 50 ppm. However, soils near ore deposits,
phosphate rock, or ore-smelting facilities or soils contaminated by airport or highway traffic or
near other source areas may contain higher concentrations (IARC 2006).
This draft document should not be construed to represent final NTP determination or policy
II-11
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
Other sources of exposure to the general public
The general public is exposed to cobalt primarily through consumption of food and to a lesser
degree through inhalation of ambient air and ingestion of drinking water; daily cobalt intake
from food has been reported to range from 5 to 50 µg/day (ATSDR 2004, Lison 2015). Although
this amount includes cobalt as part of both vitamin B12 and other cobalt compounds (ATSDR
2004), green, leafy vegetables and fresh cereals generally contain the most cobalt (IARC 1991),
and these plant sources of cobalt do not contain vitamin B12. In the 1960s, some breweries added
cobalt salts to beer to stabilize the foam (resulting in exposures of 0.04 to 0.14 mg cobalt/kg
body weight), but cobalt is no longer added to beer (ATSDR 2004). Higher cobalt intake may
result from consumption of over-the-counter or prescription mineral preparations containing
cobalt compounds.
Other potential sources of exposure include consumer products and tobacco smoking.
Cobalt is present in only a few consumer products, including cleaners, detergents, soaps, car
waxes, and a nickel metal hydride battery (5% to 10% cobalt) (ATSDR 2004, HPD 2014).
Various brands of tobacco have been reported to contain cobalt at concentrations ranging from
less than 0.3 to 2.3 µg/g dry weight, and 0.5% of the cobalt content is transferred to mainstream
smoke (WHO 2006). However, urinary cobalt levels (unadjusted for creatinine) for cigarettesmoke-exposed and unexposed NHANES participants for survey years 1999 to 2004 did not
differ significantly (Richter et al. 2009).
Regulations
Coast Guard, Department of Homeland Security
Minimum requirements have been established for safe transport of cobalt naphthenate in solvent
naphtha on ships and barges.
Department of Transportation (DOT)
Numerous cobalt compounds are considered hazardous materials, and special requirements have
been set for marking, labeling, and transporting these materials.
Environmental Protection Agency (EPA)
Clean Air Act
National Emission Standards for Hazardous Air Pollutants: Cobalt compounds are listed as
hazardous air pollutants.
Clean Water Act
Cobalt discharge limits are imposed for numerous processes during the production of cobalt at
secondary cobalt facilities processing tungsten carbide scrap raw materials.
Discharge limits for cobalt are imposed for numerous processes during the production of cobalt
at primary cobalt facilities; for numerous processes during the production of batteries; and for
numerous processes during the production of cobalt salts.
Discharge limits for cobalt are imposed for wastewater discharges from centralized waste
treatment facilities except discharges and activities exempted in 40 CFR 437.1(b), (c), and 40
CFR 421, Subpart AC.
II-12
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
Cobaltous bromide, formate, and sulfamate are designated as hazardous substances.
Comprehensive Environmental Response, Compensation, and Liability Act
Reportable quantity (RQ) = 1,000 lb for cobaltous bromide, formate, and sulfamate.
Emergency Planning and Community Right-To-Know Act
Toxics Release Inventory: Cobalt and cobalt compounds are listed substances subject to reporting
requirements.
Reportable quantity (RQ) = 100 lb for cobalt, ((2,2′-(1,2-ethanediylbis (nitrilomethylidyne))
bis(6-fluorophenolato))(2-)-N,N′,O,O′)- (also called fluomine); = 10 lb for cobalt carbonyl.
Threshold planning quantity (TPQ) = 100 lb for fluomine (solids in powder form with particle
size < 100 µm or solution or molten form); = 10,000 lb for all other forms of fluomine; = 10 lb
for cobalt carbonyl (solids in powder form with particle size < 100 µm or solution or molten
form); = 10,000 lb for all other forms of cobalt carbonyl.
Federal Insecticide, Fungicide, and Rodenticide Act
Boiled linseed oil (containing no more than 0.33% manganese naphthenate and no more than
0.33% cobalt naphthenate) is exempt from the requirement of a tolerance when used as a coating
agent for S-ethyl hexahydro-1H-azepine-1-carbothioate. No more than 15% of the pesticide
formulation may consist of boiled linseed oil, and this exemption is limited to use on rice before
edible parts form.
Food and Drug Administration (FDA)
Cobaltous salts are prohibited from use in human food.
All drugs containing cobalt salts (except radioactive forms of cobalt and its salts and cobalamin
and its derivatives) have been withdrawn from the market because they were found to be unsafe
or not effective, and they may not be compounded.
Chromium–cobalt–aluminum oxide used as a color additive for linear polyethylene surgical
sutures used in general surgery must comprise no more than 2% by weight of the suture material,
not migrate to surrounding tissue, and conform to labeling requirements in 21 CFR 70.25.
Chromium cobalt-aluminum oxide may be used as a color additive in contact lenses in amounts
not to exceed the minimum reasonably required to accomplish the intended coloring effect.
Ferric ammonium ferrocyanide and ferric ferrocyanide used to color externally applied drugs
(including those for use in the area of the eye) must not contain more than 200 ppm cobalt (as
Co) and conform to labeling requirements in 21 CFR 70.25.
21 CFR 369 contains recommended drug labeling statements for over-the-counter cobalt
preparations containing ≥ 0.5 mg cobalt as a cobalt salt per dosage unit and which recommend
administration rates of ≥ 0.5 mg per dose and ≥ 2 mg per 24-hour period.
An approved new drug application is required for marketing cobalt preparations intended for use
by man.
21 CFR 872, 874, and 888 identify class designations (Class I, II, or III) of various cobaltcontaining dental prosthetic device alloys, cobalt-chromium-alloy-based facial prosthetics, and
This draft document should not be construed to represent final NTP determination or policy
II-13
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
cobalt-chromium-molybdenum orthopedic devices that determine the type of premarketing
submission or application required for FDA clearance to market.
Cobalt naphthenate may be used in quantities that do not exceed those reasonably required as an
accelerator in the production of cross-linked polyester resins used as articles or components of
articles intended for repeated use in contact with food.
Cobalt aluminate may be safely used as a colorant in the manufacture of articles or components
of articles intended for use in producing, manufacturing, packing, processing, preparing, treating,
packaging, transporting, or holding of food at levels not to exceed 5% by weight of all polymers
except in resinous and polymeric coatings complying with 21 CFR 175.300, melamineformaldehyde resins in molded articles complying with 21 CFR 177.1460, xylene-formaldehyde
resins complying with 21 CFR 175.380, ethylene-vinyl acetate copolymers complying with 21
CFR 177.1350, and urea-formaldehyde resins in molded articles complying with 21 CFR
177.1900.
Occupational Safety and Health Administration (OSHA)
This legally enforceable PEL was adopted from the 1968 ACGIH TLV-TWA shortly after
OSHA was established; it may not reflect the most recent scientific evidence and may not
adequately protect worker health.
Permissible exposure limit (PEL) (8-h TWA) = 0.1 mg/m3 for cobalt metal, dust, and fume (as
Co).
Guidelines
American Conference of Governmental Industrial Hygienists (ACGIH)
Threshold limit value – time-weighted average (TLV-TWA) = 0.02 mg/m3 for cobalt and
inorganic compounds; = 0.1 mg/m3 for cobalt carbonyl and cobalt hydrocarbonyl.
Biological exposure index (BEI) (end of shift at end of workweek) = 15 µg/L for cobalt in urine.
Consumer Product Safety Commission (CPSC)
The CPSC has issued guidance regarding the potential hazards of specific cobalt- or cobaltcompound-containing art and craft materials (e.g., glazes, glass colorants, paints, toners,
pigments, and dyes) and specific precautions to take when using them.
Environmental Protection Agency (EPA)
Regional Screening Levels (formerly Preliminary Remediation Goals): residential soil =
23 mg/kg; industrial soil = 350 mg/kg; residential air = 0.00031 µg/m3; industrial air =
0.0014 µg/m3; tap water = 6 µg/L.
National Institute for Occupational Safety and Health (NIOSH)
Recommended exposure limit (REL) (10-h TWA) = 0.05 mg/m3 for cemented tungsten carbide
containing > 2% Co (as Co); = 0.05 mg/m3 for cobalt metal dust and fume (as Co); = 0.1 mg/m3
for cobalt carbonyl (as Co) and cobalt hydrocarbonyl (as Co).
Immediately dangerous to life and health (IDLH) limit = 20 mg/m3 for cobalt metal dust and
fume (as Co).
II-14
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
References
ATSDR. 2004. Toxicological Profile for Cobalt. Atlanta, GA: Agency for Toxic Substances and
Disease Registry. pp. 207-E203. http://www.atsdr.cdc.gov/ToxProfiles/TP.asp?id=373&tid=64.
Basu N, Abare M, Buchanan S, Cryderman D, Nam DH, Sirkin S, Schmitt S, Hu H. 2010. A
combined ecological and epidemiologic investigation of metal exposures amongst Indigenous
peoples near the Marlin Mine in Western Guatemala. Sci Total Environ 409(1): 70-77.
Behl M, Stout MD, Herbert RA, Dill JA, Baker GL, Hayden BK, Roycroft JR, Bucher JR, Hooth
MJ. 2015. Comparative toxicity and carcinogenicity of soluble and insoluble cobalt compounds.
Toxicology 333: 195-205.
Beyersmann D, Hartwig A. 2008. Carcinogenic metal compounds: recent insight into molecular
and cellular mechanisms. Arch Toxicol 82(8): 493-512.
Brock T, Stopford W. 2003. Bioaccessibility of metals in human health risk assessment:
evaluating risk from exposure to cobalt compounds. J Environ Monit 5(4): 71N-76N.
CDC. 2013. Biomonitoring Summary: Cobalt. Centers for Disease Control and Prevention.
http://www.cdc.gov/biomonitoring/Cobalt_BiomonitoringSummary.html. Last accessed:
1/31/15.
CDC. 2015. Fourth National Report on Human Exposure to Environmental Chemicals. Atlanta,
GA: Centers for Disease Control and Prevention. 514 pp.
CDI. 2006. Cobalt Facts. Cobalt Development Institute. http://thecdi.com/cobaltfacts.php. Last
accessed: 2/12/15.
ChemIDplus. 2015. ChemIDplus Advanced. National Library of Medicine.
http://chem.sis.nlm.nih.gov/chemidplus/chemidheavy.jsp and select Registry Number and search
on CAS number. Last accessed: 3/24/15.
Cheng Y, Chen G, Hong L, Zhou L, Hu M, Li B, Huang J, Xia L, Li C. 2013. How does hypoxia
inducible factor-1α participate in enhancing the glycolysis activity in cervical cancer? Ann
Diagn Pathol 17(3): 305-311.
Davis JR, ed. 2000. Nickel, Cobalt, and Their Alloys. Materials Park, OH: ASM International.
pp. 345-406.
Devlin JJ, Pomerleau AC, Brent J, Morgan BW, Deitchman S, Schwartz M. 2013. Clinical
features, testing, and management of patients with suspected prosthetic hip-associated cobalt
toxicity: a systematic review of cases. J Med Toxicol 9(4): 405-415.
Farquharson JP. 2015. Formation Metals, Inc. – President's Letter to Shareholders. Formation
Metals, Inc. Last updated: 2/23/15.
http://www.formationmetals.com/s/CobaltNews.asp?ReportID=697830.
Galán-Cobo A, Sánchez-Silva R, Serna A, Abreu-Rodríguez I, Muñoz-Cabello AM, Echevarría
M. 2013. Cellular overexpression of Aquaporins slows down the natural HIF-2α degradation
during prolonged hypoxia. Gene 522(1): 18-26.
Galanis A, Pappa A, Giannakakis A, Lanitis E, Dangaj D, Sandaltzopoulos R. 2008. Reactive
oxygen species and HIF-1 signalling in cancer. Cancer Lett 266(1): 12-20.
This draft document should not be construed to represent final NTP determination or policy
II-15
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
Galanis A, Karapetsas A, Sandaltzopoulos R. 2009. Metal-induced carcinogenesis, oxidative
stress and hypoxia signalling. Mutation Research – Genetic Toxicology and Environmental
Mutagenesis 674(1-2): 31-35.
Gao S, Zhou J, Zhao Y, Toselli P, Li W. 2013. Hypoxia-response element (HRE)-directed
transcriptional regulation of the rat lysyl oxidase gene in response to cobalt and cadmium.
Toxicol Sci 132(2): 379-389.
Gilman JP, Ruckerbauer GM. 1962. Metal carcinogenesis. I. Observations on the carcinogenicity
of a refinery dust, cobalt oxide, and colloidal thorium dioxide. Cancer Res 22: 152-157.
Greim H, Hartwig A, Reuter U, Richter-Reichhelm HB, Thielmann HW. 2009. Chemically
induced pheochromocytomas in rats: Mechanisms and relevance for human risk assessment. Crit
Rev Toxicol 39(8): 695-718.
Grimsrud TK, Berge SR, Haldorsen T, Andersen A. 2005. Can lung cancer risk among nickel
refinery workers be explained by occupational exposures other than nickel? Epidemiology 16(2):
146-154.
Hansen T, Clermont G, Alves A, Eloy R, Brochhausen C, Boutrand JP, Gatti AM, Kirkpatrick
CJ. 2006. Biological tolerance of different materials in bulk and nanoparticulate form in a rat
model: sarcoma development by nanoparticles. J R Soc Interface 3(11): 767-775.
Heath JC. 1956. The production of malignant tumours by cobalt in the rat. Br J Cancer 10(4):
668-673.
Heath JC, Daniel MR. 1962. The production of malignant tumours by cobalt in the rat:
intrathoracic tumours. Br J Cancer 16(3): 473-478.
Hillwalker WE, Anderson KA. 2014. Bioaccessibility of metals in alloys: evaluation of three
surrogate biofluids. Environ Pollut 185: 52-58.
HPD. 2014. Household Products Database. National Library of Medicine. Last updated: 8/14.
http://householdproducts.nlm.nih.gov/advancedsearch.htm and select Ingredient and search on
CAS number.
HSDB. 2015. Hazardous Substances Data Bank. National Library of Medicine.
http://toxnet.nlm.nih.gov/cgi-bin/sis/htmlgen?HSDB and search on CAS number. Last accessed:
5/27/15.
IARC. 1991. Cobalt and cobalt compounds. In Chlorinated Drinking-water; Chlorination Byproducts; Some Other Halogenated Compounds; Cobalt and Cobalt Compounds. IARC
Monographs on the Evaluation of Carcinogenic Risks to Humans, vol. 52. Lyon, France:
International Agency for Research on Cancer. pp. 363-434.
IARC. 2006. Metallic cobalt particles (with or without tungsten carbide). In Cobalt in Hard
Metals. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, vol. 86. Lyon,
France: International Agency for Research on Cancer. pp. 39-155.
Jasmin G, Riopelle JL. 1976. Renal carcinomas and erythrocytosis in rats following intrarenal
injection of nickel subsulfide. Lab Invest 35(1): 71-78.
Lison D. 2015. Cobalt. In Handbook on the Toxicology of Metals, 4th ed., vol. II. Nordberg GF,
Fowler BA, Nordberg M, eds. Waltham, MA: Elsevier. pp. 743-763.
II-16
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
Maverick T. 2015. Cobalt shortage put brakes on electric car. Wall Street Daily.
http://www.wallstreetdaily.com/2015/01/13/cobalt-electric-car-battery. Last accessed: 5/18/15.
Maxwell P, Salnikow K. 2004. HIF-1: an oxygen and metal responsive transcription factor.
Cancer Biol Ther 3(1): 29-35.
Mayo Clinic. 2015. Test ID: COS. Cobalt, Serum. Mayo Medical Laboratories.
http://www.mayomedicallaboratories.com/test-catalog/Clinical+and+Interpretive/80084. Last
accessed: 4/10/15.
MHRA. 2012. Medical Device Alert: All Metal-on-Metal (MoM) Hip Replacements.
MDA/2012/036. Medicines and Healthcare Products Regulatory Agency, National Joint Registry
of England, Wales, and Northern Ireland. https://assets.digital.cabinetoffice.gov.uk/media/5485abf6ed915d4c10000273/con155767.pdf.
Moreno ME, Acosta-Saavedra LC, Meza-Figueroa D, Vera E, Cebrian ME, Ostrosky-Wegman
P, Calderon-Aranda ES. 2010. Biomonitoring of metal in children living in a mine tailings zone
in Southern Mexico: A pilot study. Int J Hyg Environ Health 213(4): 252-258.
MorningStar. 2005a. Summary of Existing Data, Proposed Test Plan and Rationale for Cobalt
Naphthenate (CASRN 61789-51-3). U.S. High Production Volume Challenge Program. Prepared
by MorningStar Consulting, Inc., on behalf of OM Group, Inc. and the Shepherd Chemical
Company and Troy Corporation. Submitted to the U.S. Environmental Protection Agency.
http://www.epa.gov/hpv/pubs/summaries/metalcarb/c14172rt11.pdf.
MorningStar. 2005b. Summary of Existing Data, Proposed Test Plan and Rationale for Cobalt
Cobalt Salts of C8 to C13 Carboxylic Acids Including Neodecanoic Acid, Cobalt Salt (CASRN
27253-31-2), Fatty Acids, C9-C13 Neo, Cobalt Salts (CASRN 68955-83-9) and Hexanoic Acid,
2-Ethyl, Cobalt Salt (CASRN 136-52-7). U.S. High Production Volume Challenge Program.
Prepared by MorningStar Consulting, Inc., on behalf of OM Group, Inc. and the Shepherd
Chemical Company and Troy Corporation. Submitted to the U.S. Environmental Protection
Agency. http://www.epa.gov/HPV/pubs/summaries/metalcarb/c14172rt9.pdf.
Moulin JJ, Wild P, Mur JM, Fournier-Betz M, Mercier-Gallay M. 1993. A mortality study of
cobalt production workers: an extension of the follow-up. Am J Ind Med 23(2): 281-288.
Moulin JJ, Wild P, Romazini S, Lasfargues G, Peltier A, Bozec C, Deguerry P, Pellet F, Perdrix
A. 1998. Lung cancer risk in hard-metal workers. Am J Epidemiol 148(3): 241-248.
Moulin JJ, Clavel T, Roy D, Dananché B, Marquis N, Févotte J, Fontana JM. 2000. Risk of lung
cancer in workers producing stainless steel and metallic alloys. Int Arch Occup Environ Health
73(3): 171-180.
Mur JM, Moulin JJ, Charruyer-Seinerra MP, Lafitte J. 1987. A cohort mortality study among
cobalt and sodium workers in an electrochemical plant. Am J Ind Med 11(1): 75-81.
Neale G. 1990. B12 binding proteins. Gut 31(1): 59-63.
Nilsson K, Jensen BS, Carlsen L. 1985. The migration chemistry of cobalt. Eur Appl Res Rept –
Nucl Sci Technol 7(1): 23-86.
This draft document should not be construed to represent final NTP determination or policy
II-17
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
NIOSH. 1990. National Occupational Hazard Survey (1981-1983). National Institute for
Occupational Safety and Health. Last updated: 7/1/90.
http://www.cdc.gov/noes/noes4/73470sco.html, http://www.cdc.gov/noes/noes4/x8305sco.html.
NTP. 1998. Toxicology and Carcinogenesis Studies of Cobalt Sulfate Heptahydrate (CAS No.
10026-24-1) in F344/N Rats and B6C3F1/N Mice (Inhalation Studies). Technical Report Series
No. 471. NIH Publication No. 98-3961. Research Triangle Park, NC: National Toxicology
Program. 268 pp.
NTP. 2009. Report on Carcinogens Background Document for Cobalt–Tungsten Carbide:
Powders and Hard Metals. Research Triangle Park, NC: National Toxicology Program.
http://ntp.niehs.nih.gov/ntp/roc/twelfth/2010/finalbds/hardmetalsbd20100408_508.pdf.
NTP. 2014. Toxicology Studies of Cobalt Metal (CAS No. 7440-48-4) in F344/N Rats and
B6C3F1/N Mice and Toxicology and Carcinogenesis Studies of Cobalt Metal in F344/NTac Rats
and B6C3F1/N Mice (Inhalation Studies). Technical Report Series No. 581. NIH Publication No.
14-5932. Research Triangle Park, NC: National Toxicology Program. 308 pp.
Nyga A, Hart A, Tetley TD. 2015. Importance of the HIF pathway in cobalt nanoparticleinduced cytotoxicity and inflammation in human macrophages. Nanotoxicology: 1-13.
O'Rorke MA, Cantwell MM, Abnet CC, Brockman AJ, Murray LJ, Group FS. 2012. Toenail
trace element status and risk of Barrett's oesophagus and oesophageal adenocarcinoma: results
from the FINBAR study. Int J Cancer 131(8): 1882-1891.
Paul SAM, Simons JW, Mabjeesh NJ. 2004. HIF at the crossroads between ischemia and
carcinogenesis. J Cell Physiol 200(1): 20-30.
Paustenbach DJ, Tvermoes BE, Unice KM, Finley BL, Kerger BD. 2013. A review of the health
hazards posed by cobalt. Crit Rev Toxicol 43(4): 316-362.
Peters K, Unger RE, Gatti AM, Sabbioni E, Tsaryk R, Kirkpatrick CJ. 2007. Metallic
nanoparticles exhibit paradoxical effects on oxidative stress and pro-inflammatory response in
endothelial cells in vitro. Int J Immunopathol Pharmacol 20(4): 679-689.
PubChem. 2015. PubChem Compound Database. http://www.ncbi.nlm.nih.gov/pccompound.
Last accessed: 5/27/15.
Richter PA, Bishop EE, Wang J, Swahn MH. 2009. Tobacco smoke exposure and levels of
urinary metals in the U.S. youth and adult population: the National Health and Nutrition
Examination Survey (NHANES) 1999-2004. Int J Environ Res Public Health 6(7): 1930-1946.
Rogers MA, Thomas DB, Davis S, Vaughan TL, Nevissi AE. 1993. A case-control study of
element levels and cancer of the upper aerodigestive tract. Cancer Epidemiol Biomarkers Prev
2(4): 305-312.
Sabbioni E, Fortaner S, Farina M, Del Torchio R, Petrarca C, Bernardini G, et al. 2014.
Interaction with culture medium components, cellular uptake and intracellular distribution of
cobalt nanoparticles, microparticles and ions in Balb/3T3 mouse fibroblasts. Nanotoxicology
8(1): 88-99.
II-18
This draft document should not be construed to represent final NTP determination or policy
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
6/05/15
Saini Y, Greenwood KK, Merrill C, Kim KY, Patial S, Parameswaran N, Harkema JR, LaPres
JJ. 2010a. Acute cobalt-induced lung injury and the role of hypoxia-inducible factor 1α in
modulating inflammation. Toxicol Sci 116(2): 673-681.
Saini Y, Kim KY, Lewandowski R, Bramble LA, Harkema JR, Lapres JJ. 2010b. Role of
hypoxia-inducible factor 1α in modulating cobalt-induced lung inflammation. Am J Physiol
Lung Cell Mol Physiol 298(2): L139-147.
Sampson B, Hart A. 2012. Clinical usefulness of blood metal measurements to assess the failure
of metal-on-metal hip implants. Ann Clin Biochem 49(Pt 2): 118-131.
Shabaan AA, Marks V, Lancaster MC, Dufeu GN. 1977. Fibrosarcomas induced by cobalt
chloride (CoCl2) in rats. Lab Anim 11(1): 43-46.
Shedd KB. 1993. The Materials Flow of Cobalt in the United States. Bureau of Mines
Information Circular 9350. United States Department of the Interior.
http://pubs.usgs.gov/usbmic/ic-9350/ic-9350.pdf.
Shedd KB. 2014a. Cobalt. In Mineral Commodity Summaries 2014. U.S. Geological Survey.
http://minerals.usgs.gov/minerals/pubs/commodity/cobalt/mcs-2014-cobal.pdf.
Shedd KB. 2014b. USGS 2012 Minerals Yearbook: Cobalt [Advance Release].
http://minerals.usgs.gov/minerals/pubs/commodity/cobalt/myb1-2012-cobal.pdf.
Smith LJ, Holmes AL, Kandpal SK, Mason MD, Zheng T, Wise JP, Sr. 2014. The cytotoxicity
and genotoxicity of soluble and particulate cobalt in human lung fibroblast cells. Toxicol Appl
Pharmacol 278(3): 259-265.
Steinhoff D, Mohr U. 1991. On the question of a carcinogenic action of cobalt-containing
compounds. Exp Pathol 41(4): 169-174.
Stopford W, Turner J, Cappellini D, Brock T. 2003. Bioaccessibility testing of cobalt
compounds. J Environ Monit 5(4): 675-680.
TRI. 2014a. TRI Explorer Chemical Report. U.S. Environmental Protection Agency.
http://www.epa.gov/triexplorer. Last accessed: 10/14/14.
TRI. 2014b. TRI EZ Search. Envirofacts. U.S. Environmental Protection Agency.
http://www.epa.gov/enviro/facts/tri/ez.html. Last accessed: 10/29/14.
Tüchsen F, Jensen MV, Villadsen E, Lynge E. 1996. Incidence of lung cancer among cobaltexposed women. Scand J Work Environ Health 22(6): 444-450.
Valko M, Morris H, Cronin MT. 2005. Metals, toxicity and oxidative stress. Curr Med Chem
12(10): 1161-1208.
Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M. 2006. Free radicals, metals and
antioxidants in oxidative stress-induced cancer. Chem Biol Interact 160(1): 1-40.
Wehner AP, Busch RH, Olson RJ, Craig DK. 1977. Chronic inhalation of cobalt oxide and
cigarette smoke by hamsters. Am Ind Hyg Assoc J 38(7): 338-346.
WHO. 2006. Concise International Chemical Assessment Document 69. Cobalt and Inorganic
Cobalt Compounds. Geneva: World Health Organization.
http://www.inchem.org/documents/cicads/cicads/cicad69.htm.
This draft document should not be construed to represent final NTP determination or policy
II-19
6/05/15
Peer-Review Draft: Profile on Cobalt and Certain Cobalt Compounds
Wild P, Perdrix A, Romazini S, Moulin JJ, Pellet F. 2000. Lung cancer mortality in a site
producing hard metals. Occup Environ Med 57(8): 568-573.
II-20
This draft document should not be construed to represent final NTP determination or policy